langgraph 0.2.40

Add "node finished" callback

.github/scripts/run_langgraph_cli_test.py

@@ -0,0 +1,116 @@
+import asyncio
+import json
+import os
+import pathlib
+import sys
+import langgraph_cli
+import langgraph_cli.docker
+import langgraph_cli.config
+
+from langgraph_cli.exec import Runner, subp_exec
+from langgraph_cli.progress import Progress
+from langgraph_cli.constants import DEFAULT_PORT
+
+
+def test(
+    config: pathlib.Path,
+    port: int,
+    tag: str,
+    verbose: bool,
+):
+    with Runner() as runner, Progress(message="Pulling...") as set:
+        # check docker available
+        capabilities = langgraph_cli.docker.check_capabilities(runner)
+        # open config
+        with open(config) as f:
+            config_json = langgraph_cli.config.validate_config(json.load(f))
+
+        set("Running...")
+        args = [
+            "run",
+            "--rm",
+            "-p",
+            f"{port}:8000",
+        ]
+        if isinstance(config_json["env"], str):
+            args.extend(
+                [
+                    "--env-file",
+                    str(config.parent / config_json["env"]),
+                ]
+            )
+        else:
+            for k, v in config_json["env"].items():
+                args.extend(
+                    [
+                        "-e",
+                        f"{k}={v}",
+                    ]
+                )
+        if capabilities.healthcheck_start_interval:
+            args.extend(
+                [
+                    "--health-interval",
+                    "5s",
+                    "--health-retries",
+                    "1",
+                    "--health-start-period",
+                    "10s",
+                    "--health-start-interval",
+                    "1s",
+                ]
+            )
+        else:
+            args.extend(
+                [
+                    "--health-interval",
+                    "5s",
+                    "--health-retries",
+                    "2",
+                ]
+            )
+
+        _task = None
+
+        def on_stdout(line: str):
+            nonlocal _task
+            if "GET /ok" in line or "Uvicorn running on" in line:
+                set("")
+                sys.stdout.write(
+                    f"""Ready!
+- API: http://localhost:{port}
+"""
+                )
+                sys.stdout.flush()
+                _task.cancel()
+                return True
+            return False
+
+        async def subp_exec_task(*args, **kwargs):
+            nonlocal _task
+            _task = asyncio.create_task(subp_exec(*args, **kwargs))
+            await _task
+
+        try:
+            runner.run(
+                subp_exec_task(
+                    "docker",
+                    *args,
+                    tag,
+                    verbose=verbose,
+                    on_stdout=on_stdout,
+                )
+            )
+        except asyncio.CancelledError:
+            pass
+
+
+if __name__ == "__main__":
+    import argparse
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("-t", "--tag", type=str)
+    parser.add_argument("-c", "--config", type=str, default="./langgraph.json")
+    parser.add_argument("-p", "--port", default=DEFAULT_PORT)
+    args = parser.parse_args()
+    test(pathlib.Path(args.config), args.port, args.tag, verbose=True)

.github/workflows/_integration_test.yml

@@ -39,22 +39,36 @@ jobs:
       - name: Install cli globally
         if: steps.changed-files.outputs.all
         run: pip install -e .
-      - name: Start service A
+      - name: Build and test service A
         if: steps.changed-files.outputs.all
+        working-directory: libs/cli/examples
         run: |
-          timeout 60 langgraph test -c examples/langgraph.json --verbose || (exit "$(($? == 124 ? 0 : $?))")
-      - name: Start service B
+          # The build-arg isn't used; just testing that we accept other args
+          langgraph build -t langgraph-test-a  --base-image "langchain/langgraph-trial"
+          cp .env.example .envg
+          timeout 60 python ../../../.github/scripts/run_langgraph_cli_test.py -c langgraph.json -t langgraph-test-a
+      - name: Build and test service B
         if: steps.changed-files.outputs.all
         working-directory: libs/cli/examples/graphs
         run: |
-          timeout 60 langgraph test --verbose || (exit "$(($? == 124 ? 0 : $?))")
-      - name: Start service C
+          langgraph build -t langgraph-test-b  --base-image "langchain/langgraph-trial"
+          timeout 60 python ../../../../.github/scripts/run_langgraph_cli_test.py -t langgraph-test-b
+      - name: Build and test service C
         if: steps.changed-files.outputs.all
         working-directory: libs/cli/examples/graphs_reqs_a
         run: |
-          timeout 60 langgraph test --verbose || (exit "$(($? == 124 ? 0 : $?))")
-      - name: Start service D
+          langgraph build -t langgraph-test-c  --base-image "langchain/langgraph-trial"
+          timeout 60 python ../../../../.github/scripts/run_langgraph_cli_test.py -t langgraph-test-c
+      - name: Build and test service D
         if: steps.changed-files.outputs.all
         working-directory: libs/cli/examples/graphs_reqs_b
         run: |
-          timeout 60 langgraph test --verbose || (exit "$(($? == 124 ? 0 : $?))")
+          langgraph build -t langgraph-test-d  --base-image "langchain/langgraph-trial"
+          timeout 60 python ../../../../.github/scripts/run_langgraph_cli_test.py -t langgraph-test-d
+
+      - name: Build JS service
+        if: steps.changed-files.outputs.all
+        working-directory: libs/cli/js-examples
+        run: |
+          langgraph build -t langgraph-test-e
+  

.github/workflows/_test.yml

@@ -21,6 +21,7 @@ jobs:
           - "3.10"
           - "3.11"
           - "3.12"
+          - "3.13"
 
     name: "test #${{ matrix.python-version }}"
     steps:
@@ -32,6 +33,12 @@ jobs:
           poetry-version: ${{ env.POETRY_VERSION }}
           working-directory: ${{ inputs.working-directory }}
           cache-key: test-${{ inputs.working-directory }}
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        if: ${{ !github.event.pull_request.head.repo.fork }}
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_RO_TOKEN }}
 
       - name: Install dependencies
         shell: bash

.github/workflows/_test_langgraph.yml

@@ -16,9 +16,12 @@ jobs:
           - "3.10"
           - "3.11"
           - "3.12"
+          - "3.13"
         core-version:
-          - ">=0.2.39,<0.3.0"
           - "latest"
+        include:
+          - python-version: "3.11"
+            core-version: ">=0.2.39,<0.3.0"
 
     defaults:
       run:
@@ -32,6 +35,12 @@ jobs:
           python-version: ${{ matrix.python-version }}
           poetry-version: ${{ env.POETRY_VERSION }}
           cache-key: test-langgraph
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        if: ${{ !github.event.pull_request.head.repo.fork }}
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_RO_TOKEN }}
 
       - name: Install dependencies
         shell: bash

.github/workflows/_test_scheduler_kafka.yml

@@ -27,6 +27,12 @@ jobs:
           python-version: ${{ matrix.python-version }}
           poetry-version: ${{ env.POETRY_VERSION }}
           cache-key: test-scheduler-kafka
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        if: ${{ !github.event.pull_request.head.repo.fork }}
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_RO_TOKEN }}
 
       - name: Install dependencies
         shell: bash

.github/workflows/ci.yml

@@ -31,6 +31,7 @@ jobs:
             "libs/cli",
             "libs/checkpoint",
             "libs/checkpoint-sqlite",
+            "libs/checkpoint-duckdb",
             "libs/checkpoint-postgres",
             "libs/scheduler-kafka",
           ]
@@ -47,6 +48,7 @@ jobs:
             "libs/cli",
             "libs/checkpoint",
             "libs/checkpoint-sqlite",
+            "libs/checkpoint-duckdb",
             "libs/checkpoint-postgres"
           ]
     uses: ./.github/workflows/_test.yml

.github/workflows/deploy_docs.yml

@@ -58,7 +58,7 @@ jobs:
 
       - name: Install dependencies
         run: |
-          poetry install --with docs
+          poetry install --with test
           poetry run pip install -U pytest pytest-check-links langsmith langchain GitPython
 
       - name: Lint Docs
@@ -81,7 +81,10 @@ jobs:
               --check-links-ignore "https://x.com/.*" \
               --check-links-ignore "https://github\.com/.*" \
               --check-links-ignore "/.*\.(ipynb|html)$" \
-              --check-links $(find docs/site -name "index.html" | grep -v 'storm/index.html')
+              --check-links-ignore "https://python\.langchain\.com/.*" \
+              --check-links-ignore "https://openai.com/index/memory-and-new-controls-for-chatgpt/" \
+              --check-links $(find docs/site -name "index.html" | grep -v 'storm/index.html') 
+              
           else
             echo "Fetching changes from origin/main..."
             git fetch origin main

docs/_scripts/execute_notebooks.sh

@@ -24,7 +24,7 @@ export -f execute_notebook
 
 # Check if custom notebook paths are provided
 if [ $# -gt 0 ]; then
-    notebooks="$@"
+    notebooks=$(echo "$@" | tr ' ' '\n' | grep -vFf <(echo "$SKIP_NOTEBOOKS"))
 else
     # Find all notebooks and filter out those in the skip list
     notebooks=$(find docs/docs/tutorials docs/docs/how-tos -name "*.ipynb" | grep -v ".ipynb_checkpoints" | grep -vFf <(echo "$SKIP_NOTEBOOKS"))

docs/_scripts/generate_api_reference_links.py

@@ -0,0 +1,246 @@
+import importlib
+import inspect
+import logging
+import os
+import re
+from typing import List, Literal, Optional
+from typing_extensions import TypedDict
+
+import nbformat
+from nbconvert.preprocessors import Preprocessor
+
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+# Base URL for all class documentation
+_LANGCHAIN_API_REFERENCE = "https://python.langchain.com/api_reference/"
+_LANGGRAPH_API_REFERENCE = "https://langchain-ai.github.io/langgraph/reference/"
+
+
+# (alias/re-exported modules, source module, class, docs namespace)
+MANUAL_API_REFERENCES_LANGGRAPH = [
+    (
+        ["langgraph.prebuilt"],
+        "langgraph.prebuilt.chat_agent_executor",
+        "create_react_agent",
+        "prebuilt",
+    ),
+    (["langgraph.prebuilt"], "langgraph.prebuilt.tool_node", "ToolNode", "prebuilt"),
+    (
+        ["langgraph.prebuilt"],
+        "langgraph.prebuilt.tool_node",
+        "tools_condition",
+        "prebuilt",
+    ),
+    (
+        ["langgraph.prebuilt"],
+        "langgraph.prebuilt.tool_node",
+        "InjectedState",
+        "prebuilt",
+    ),
+    # Graph
+    (["langgraph.graph"], "langgraph.graph.message", "add_messages", "graphs"),
+    (["langgraph.graph"], "langgraph.graph.state", "StateGraph", "graphs"),
+    (["langgraph.graph"], "langgraph.graph.state", "CompiledStateGraph", "graphs"),
+    ([], "langgraph.types", "StreamMode", "types"),
+    (["langgraph.graph"], "langgraph.constants", "START", "constants"),
+    (["langgraph.graph"], "langgraph.constants", "END", "constants"),
+    (["langgraph.constants"], "langgraph.types", "Send", "types"),
+    (["langgraph.constants"], "langgraph.types", "Interrupt", "types"),
+    ([], "langgraph.types", "RetryPolicy", "types"),
+    ([], "langgraph.checkpoint.base", "Checkpoint", "checkpoints"),
+    ([], "langgraph.checkpoint.base", "CheckpointMetadata", "checkpoints"),
+    ([], "langgraph.checkpoint.base", "BaseCheckpointSaver", "checkpoints"),
+    ([], "langgraph.checkpoint.base", "SerializerProtocol", "checkpoints"),
+    ([], "langgraph.checkpoint.serde.jsonplus", "JsonPlusSerializer", "checkpoints"),
+    ([], "langgraph.checkpoint.memory", "MemorySaver", "checkpoints"),
+    ([], "langgraph.checkpoint.sqlite.aio", "AsyncSqliteSaver", "checkpoints"),
+    ([], "langgraph.checkpoint.sqlite", "SqliteSaver", "checkpoints"),
+    ([], "langgraph.checkpoint.postgres.aio", "AsyncPostgresSaver", "checkpoints"),
+    ([], "langgraph.checkpoint.postgres", "PostgresSaver", "checkpoints"),
+]
+
+WELL_KNOWN_LANGGRAPH_OBJECTS = {
+    (module_, class_): (source_module, namespace)
+    for (modules, source_module, class_, namespace) in MANUAL_API_REFERENCES_LANGGRAPH
+    for module_ in modules + [source_module]
+}
+
+
+def _make_regular_expression(pkg_prefix: str) -> re.Pattern:
+    if not pkg_prefix.isidentifier():
+        raise ValueError(f"Invalid package prefix: {pkg_prefix}")
+    return re.compile(
+        r"from\s+(" + pkg_prefix + "(?:_\w+)?(?:\.\w+)*?)\s+import\s+"
+        r"((?:\w+(?:,\s*)?)*"  # Match zero or more words separated by a comma+optional ws
+        r"(?:\s*\(.*?\))?)",  # Match optional parentheses block
+        re.DOTALL,  # Match newlines as well
+    )
+
+
+# Regular expression to match langchain import lines
+_IMPORT_LANGCHAIN_RE = _make_regular_expression("langchain")
+_IMPORT_LANGGRAPH_RE = _make_regular_expression("langgraph")
+
+
+def _get_full_module_name(module_path, class_name) -> Optional[str]:
+    """Get full module name using inspect"""
+    try:
+        module = importlib.import_module(module_path)
+        class_ = getattr(module, class_name)
+        module = inspect.getmodule(class_)
+        if module is None:
+            # For constants, inspect.getmodule() might return None
+            # In this case, we'll return the original module_path
+            return module_path
+        return module.__name__
+    except AttributeError as e:
+        logger.warning(f"Could not find module for {class_name}, {e}")
+        return None
+    except ImportError as e:
+        logger.warning(f"Failed to load for class {class_name}, {e}")
+        return None
+
+
+def _get_doc_title(data: str, file_name: str) -> str:
+    try:
+        return re.findall(r"^#\s*(.*)", data, re.MULTILINE)[0]
+    except IndexError:
+        pass
+    # Parse the rst-style titles
+    try:
+        return re.findall(r"^(.*)\n=+\n", data, re.MULTILINE)[0]
+    except IndexError:
+        return file_name
+
+
+class ImportInformation(TypedDict):
+    imported: str  # imported class name
+    source: str  # module path
+    docs: str  # URL to the documentation
+    title: str  # Title of the document
+
+
+def _get_imports(
+    code: str, doc_title: str, package_ecosystem: Literal["langchain", "langgraph"]
+) -> List[ImportInformation]:
+    """Get imports from the given code block.
+
+    Args:
+        code: Python code block from which to extract imports
+        doc_title: Title of the document
+        package_ecosystem: "langchain" or "langgraph". The two live in different
+            repositories and have separate documentation sites.
+
+    Returns:
+        List of import information for the given code block
+    """
+    imports = []
+
+    if package_ecosystem == "langchain":
+        pattern = _IMPORT_LANGCHAIN_RE
+    elif package_ecosystem == "langgraph":
+        pattern = _IMPORT_LANGGRAPH_RE
+    else:
+        raise ValueError(f"Invalid package ecosystem: {package_ecosystem}")
+
+    for import_match in pattern.finditer(code):
+        module = import_match.group(1)
+        if "pydantic_v1" in module:
+            continue
+        imports_str = (
+            import_match.group(2).replace("(\n", "").replace("\n)", "")
+        )  # Handle newlines within parentheses
+        # remove any newline and spaces, then split by comma
+        imported_classes = [
+            imp.strip()
+            for imp in re.split(r",\s*", imports_str.replace("\n", ""))
+            if imp.strip()
+        ]
+        for class_name in imported_classes:
+            module_path = _get_full_module_name(module, class_name)
+            if not module_path:
+                continue
+            if len(module_path.split(".")) < 2:
+                continue
+
+            if package_ecosystem == "langchain":
+                pkg = module_path.split(".")[0].replace("langchain_", "")
+                top_level_mod = module_path.split(".")[1]
+
+                url = (
+                    _LANGCHAIN_API_REFERENCE
+                    + pkg
+                    + "/"
+                    + top_level_mod
+                    + "/"
+                    + module_path
+                    + "."
+                    + class_name
+                    + ".html"
+                )
+            elif package_ecosystem == "langgraph":
+                if (module, class_name) not in WELL_KNOWN_LANGGRAPH_OBJECTS:
+                    # Likely not documented yet
+                    continue
+
+                source_module, namespace = WELL_KNOWN_LANGGRAPH_OBJECTS[
+                    (module, class_name)
+                ]
+                url = (
+                    _LANGGRAPH_API_REFERENCE
+                    + namespace
+                    + "/#"
+                    + source_module
+                    + "."
+                    + class_name
+                )
+            else:
+                raise ValueError(f"Invalid package ecosystem: {package_ecosystem}")
+
+            # Add the import information to our list
+            imports.append(
+                {
+                    "imported": class_name,
+                    "source": module,
+                    "docs": url,
+                    "title": doc_title,
+                }
+            )
+
+    return imports
+
+
+class ImportPreprocessor(Preprocessor):
+    """A preprocessor to replace imports in each Python code cell with links to their
+    documentation and append the import info in a comment."""
+
+    def preprocess(self, nb, resources):
+        self.all_imports = []
+        file_name = os.path.basename(resources.get("metadata", {}).get("name", ""))
+        _DOC_TITLE = _get_doc_title(nb.cells[0].source, file_name)
+
+        cells = []
+        for cell in nb.cells:
+            if cell.cell_type == "code":
+                cells.append(cell)
+                imports = _get_imports(
+                    cell.source, _DOC_TITLE, "langchain"
+                ) + _get_imports(cell.source, _DOC_TITLE, "langgraph")
+                if not imports:
+                    continue
+
+                cells.append(
+                    nbformat.v4.new_markdown_cell(
+                        source=f"""
+<div>
+<b>API Reference:</b>
+{' | '.join(f'<a href="{imp["docs"]}">{imp["imported"]}</a>' for imp in imports)}
+</div>
+                    """
+                    )
+                )
+            else:
+                cells.append(cell)
+        nb.cells = cells
+        return nb, resources

docs/_scripts/notebook_convert.py

@@ -0,0 +1,126 @@
+import os
+import re
+from pathlib import Path
+
+import nbformat
+from nbconvert.exporters import MarkdownExporter
+from nbconvert.preprocessors import Preprocessor
+
+from generate_api_reference_links import ImportPreprocessor
+
+
+class EscapePreprocessor(Preprocessor):
+    def preprocess_cell(self, cell, resources, cell_index):
+        if cell.cell_type == "markdown":
+            # rewrite markdown links to html links (excluding image links)
+            cell.source = re.sub(
+                r"(?<!!)\[([^\]]*)\]\((?![^\)]*//)([^)]*)(?:\.ipynb)?\)",
+                r'<a href="\2">\1</a>',
+                cell.source,
+            )
+            # Fix image paths in <img> tags
+            cell.source = re.sub(
+                r'<img\s+src="\.?/img/([^"]+)"', r'<img src="../img/\1"', cell.source
+            )
+
+        elif cell.cell_type == "code":
+            # escape ``` in code
+            cell.source = cell.source.replace("```", r"\`\`\`")
+            # escape ``` in output
+            if "outputs" in cell:
+                filter_out = set()
+                for i, output in enumerate(cell["outputs"]):
+                    if "text" in output:
+                        if not output["text"].strip():
+                            filter_out.add(i)
+                            continue
+
+                        value = output["text"].replace("```", r"\`\`\`")
+                        # handle a funky case w/ references in text
+                        value = re.sub(r"\[(\d+)\](?=\[(\d+)\])", r"[\1]\\", value)
+                        output["text"] = value
+                    elif "data" in output:
+                        for key, value in output["data"].items():
+                            if isinstance(value, str):
+                                value = value.replace("```", r"\`\`\`")
+                                # handle a funky case w/ references in text
+                                output["data"][key] = re.sub(
+                                    r"\[(\d+)\](?=\[(\d+)\])", r"[\1]\\", value
+                                )
+                cell["outputs"] = [
+                    output
+                    for i, output in enumerate(cell["outputs"])
+                    if i not in filter_out
+                ]
+
+        return cell, resources
+
+
+class ExtractAttachmentsPreprocessor(Preprocessor):
+    """
+    Extracts all of the outputs from the notebook file.  The extracted
+    outputs are returned in the 'resources' dictionary.
+    """
+
+    def preprocess_cell(self, cell, resources, cell_index):
+        """
+        Apply a transformation on each cell,
+        Parameters
+        ----------
+        cell : NotebookNode cell
+            Notebook cell being processed
+        resources : dictionary
+            Additional resources used in the conversion process.  Allows
+            preprocessors to pass variables into the Jinja engine.
+        cell_index : int
+            Index of the cell being processed (see base.py)
+        """
+
+        # Get files directory if it has been specified
+
+        # Make sure outputs key exists
+        if not isinstance(resources["outputs"], dict):
+            resources["outputs"] = {}
+
+        # Loop through all of the attachments in the cell
+        for name, attach in cell.get("attachments", {}).items():
+            for mime, data in attach.items():
+                if mime not in {
+                    "image/png",
+                    "image/jpeg",
+                    "image/svg+xml",
+                    "application/pdf",
+                }:
+                    continue
+
+                # attachments are pre-rendered. Only replace markdown-formatted
+                # images with the following logic
+                attach_str = f"({name})"
+                if attach_str in cell.source:
+                    data = f"(data:{mime};base64,{data})"
+                    cell.source = cell.source.replace(attach_str, data)
+
+        return cell, resources
+
+
+exporter = MarkdownExporter(
+    preprocessors=[
+        EscapePreprocessor,
+        ExtractAttachmentsPreprocessor,
+        ImportPreprocessor,
+    ],
+    template_name="mdoutput",
+    extra_template_basedirs=[
+        os.path.join(os.path.dirname(__file__), "notebook_convert_templates")
+    ],
+)
+
+
+def convert_notebook(
+    notebook_path: Path,
+) -> Path:
+    with open(notebook_path) as f:
+        nb = nbformat.read(f, as_version=4)
+
+    body, _ = exporter.from_notebook_node(nb)
+    return body

docs/_scripts/notebook_convert_templates/mdoutput/conf.json

@@ -0,0 +1,5 @@
+{
+  "mimetypes": {
+    "text/markdown": true
+  }
+}

docs/_scripts/notebook_convert_templates/mdoutput/index.md.j2

@@ -0,0 +1,33 @@
+{% extends 'markdown/index.md.j2' %}
+
+{%- block traceback_line -%}
+```output
+{{ line.rstrip() | strip_ansi }}
+```
+{%- endblock traceback_line -%}
+
+{%- block stream -%}
+```output
+{{ output.text.rstrip() }}
+```
+{%- endblock stream -%}
+
+{%- block data_text scoped -%}
+```output
+{{ output.data['text/plain'].rstrip() }}
+```
+{%- endblock data_text -%}
+
+{%- block data_html scoped -%}
+```html
+{{ output.data['text/html'] | safe }} 
+```
+{%- endblock data_html -%}
+
+{%- block data_jpg scoped -%}
+![](data:image/jpg;base64,{{ output.data['image/jpeg'] }})
+{%- endblock data_jpg -%}
+
+{%- block data_png scoped -%}
+![](data:image/png;base64,{{ output.data['image/png'] }})
+{%- endblock data_png -%}

docs/_scripts/notebook_hooks.py

@@ -0,0 +1,40 @@
+import logging
+from typing import Any, Dict
+
+from mkdocs.structure.pages import Page
+from mkdocs.structure.files import Files, File
+from notebook_convert import convert_notebook
+
+logger = logging.getLogger(__name__)
+logging.basicConfig()
+logger.setLevel(logging.INFO)
+
+
+class NotebookFile(File):
+    def is_documentation_page(self):
+        return True
+
+
+def on_files(files: Files, **kwargs: Dict[str, Any]):
+    new_files = Files([])
+    for file in files:
+        if file.src_path.endswith(".ipynb"):
+            new_file = NotebookFile(
+                path=file.src_path,
+                src_dir=file.src_dir,
+                dest_dir=file.dest_dir,
+                use_directory_urls=file.use_directory_urls,
+            )
+            new_files.append(new_file)
+        else:
+            new_files.append(file)
+    return new_files
+
+
+def on_page_markdown(markdown: str, page: Page, **kwargs: Dict[str, Any]):
+    if page.file.src_path.endswith(".ipynb"):
+        logger.info("Processing Jupyter notebook: %s", page.file.src_path)
+        body = convert_notebook(page.file.abs_src_path)
+        return body
+
+    return markdown

docs/_scripts/prepare_notebooks_for_ci.py

@@ -40,7 +40,6 @@ NOTEBOOKS_NO_EXECUTION = [
     "docs/docs/tutorials/storm/storm.ipynb",  # issues only when running with VCR
     "docs/docs/tutorials/lats/lats.ipynb",  # issues only when running with VCR
     "docs/docs/tutorials/multi_agent/hierarchical_agent_teams.ipynb",  # taking a very long time to run
-    "docs/docs/tutorials/customer-support/customer-support.ipynb",  # user input - update
     "docs/docs/tutorials/rag/langgraph_crag.ipynb",  # flakiness from tavily
     "docs/docs/tutorials/rag/langgraph_adaptive_rag.ipynb",  # Cannot create a consistent method resolution error from VCR
     "docs/docs/how-tos/map-reduce.ipynb"  # flakiness from structured output, only when running with VCR
@@ -70,10 +69,13 @@ def is_comment(code: str) -> bool:
     return code.strip().startswith("#")
 
 
-def has_blocklisted_command(code: str) -> bool:
+def has_blocklisted_command(code: str, metadata: dict) -> bool:
+    if 'hide_from_vcr' in metadata:
+        return True
+    
     code = code.strip()
-    for blocklisted_command in BLOCKLIST_COMMANDS:
-        if blocklisted_command in code:
+    for blocklisted_pattern in BLOCKLIST_COMMANDS:
+        if blocklisted_pattern in code:
             return True
     return False
 
@@ -108,7 +110,7 @@ def add_vcr_to_notebook(
         if all(is_comment(line) or not line.strip() for line in lines):
             continue
 
-        if has_blocklisted_command(cell.source):
+        if has_blocklisted_command(cell.source, cell.metadata):
             continue
 
         cell_id = cell.get("id", idx)
@@ -125,6 +127,8 @@ def add_vcr_to_notebook(
         "import msgpack",
         "import base64",
         "import zlib",
+        "import os",
+        "os.environ.pop(\"LANGCHAIN_TRACING_V2\", None)",
         "custom_vcr = vcr.VCR()",
         "",
         "def compress_data(data, compression_level=9):",

docs/docs/cloud/deployment/cloud.md

@@ -11,7 +11,7 @@ LangGraph Cloud is available within <a href="https://www.langchain.com/langsmith
 
 Starting from the <a href="https://smith.langchain.com/" target="_blank">LangSmith UI</a>...
 
-1. In the left-hand navigation panel, select `Deployments`. The `Deployments` view contains a list of existing LangGraph Cloud deployments.
+1. In the left-hand navigation panel, select `LangGraph Cloud`. The `LangGraph Cloud` view contains a list of existing LangGraph Cloud deployments.
 1. In the top-right corner, select `+ New Deployment` to create a new deployment.
 1. In the `Create New Deployment` panel, fill out the required fields.
     1. `Deployment details`
@@ -38,7 +38,7 @@ When [creating a new deployment](#create-new-deployment), a new revision is crea
 
 Starting from the <a href="https://smith.langchain.com/" target="_blank">LangSmith UI</a>...
 
-1. In the left-hand navigation panel, select `Deployments`. The `Deployments` view contains a list of existing LangGraph Cloud deployments.
+1. In the left-hand navigation panel, select `LangGraph Cloud`. The `LangGraph Cloud` view contains a list of existing LangGraph Cloud deployments.
 1. Select an existing deployment to create a new revision for.
 1. In the `Deployment` view, in the top-right corner, select `+ New Revision`.
 1. In the `New Revision` modal, fill out the required fields.
@@ -56,7 +56,7 @@ Starting from the <a href="https://smith.langchain.com/" target="_blank">LangSmi
 
 Build and deployment logs are available for each revision.
 
-Starting from the `Deployment` view...
+Starting from the `LangGraph Cloud` view...
 
 1. Select the desired revision from the `Revisions` table. A panel slides open from the right-hand side and the `Build` tab is selected by default, which displays build logs for the revision.
 1. In the panel, select the `Deploy` tab to view deployment logs for the revision.
@@ -69,7 +69,7 @@ Interrupting a revision will stop deployment of the revision.
 !!! warning "Undefined Behavior"
     Interrupted revisions have undefined behavior. This is only useful if you need to deploy a new revision and you already have a revision "stuck" in progress. In the future, this feature may be removed.
 
-Starting from the `Deployment` view...
+Starting from the `LangGraph Cloud` view...
 
 1. Select the menu icon (three dots) on the right-hand side of the row for the desired revision from the `Revisions` table.
 1. Select `Interrupt` from the menu.
@@ -79,13 +79,13 @@ Starting from the `Deployment` view...
 
 Starting from the <a href="https://smith.langchain.com/" target="_blank">LangSmith UI</a>...
 
-1. In the left-hand navigation panel, select `Deployments`. The `Deployments` view contains a list of existing LangGraph Cloud deployments.
+1. In the left-hand navigation panel, select `LangGraph Cloud`. The `LangGraph Cloud` view contains a list of existing LangGraph Cloud deployments.
 1. Select the menu icon (three dots) on the right-hand side of the row for the desired deployment and select `Delete`.
 1. A `Confirmation` modal will appear. Select `Delete`.
 
 ## Deployment Settings
 
-Starting from the `Deployment` view...
+Starting from the `LangGraph Cloud` view...
 
 1. In the top-right corner, select the gear icon (`Deployment Settings`).
 1. Update the `Git Branch` to the desired branch.

docs/docs/cloud/how-tos/human_in_the_loop_edit_state.md

@@ -6,7 +6,7 @@ This can be in several ways, but the primary supported way is to add an "interru
 
 ## Setup
 
-We are not going to show the full code for the graph we are hosting, but you can see it [here](../../how-tos/human_in_the_loop/edit-graph-state.ipynb#build-the-agent) if you want to. Once this graph is hosted, we are ready to invoke it and wait for user input. 
+We are not going to show the full code for the graph we are hosting, but you can see it [here](../../how-tos/human_in_the_loop/edit-graph-state.ipynb#agent) if you want to. Once this graph is hosted, we are ready to invoke it and wait for user input.
 
 ### SDK initialization
 

docs/docs/cloud/how-tos/human_in_the_loop_user_input.md

@@ -14,7 +14,7 @@ Luckily, LangGraph makes it possible to do similar things in a production way. T
 
 ## Setup
 
-We are not going to show the full code for the graph we are hosting, but you can see it [here](../../how-tos/human_in_the_loop/wait-user-input.ipynb#build-the-agent) if you want to. Once this graph is hosted, we are ready to invoke it and wait for user input. 
+We are not going to show the full code for the graph we are hosting, but you can see it [here](../../how-tos/human_in_the_loop/wait-user-input.ipynb#agent) if you want to. Once this graph is hosted, we are ready to invoke it and wait for user input.
 
 ### SDK initialization
 

docs/docs/cloud/how-tos/test_deployment.md

@@ -4,7 +4,7 @@ The LangGraph Studio UI connects directly to LangGraph Cloud deployments.
 
 Starting from the <a href="https://smith.langchain.com/" target="_blank">LangSmith UI</a>...
 
-1. In the left-hand navigation panel, select `Deployments`. The `Deployments` view contains a list of existing LangGraph Cloud deployments.
+1. In the left-hand navigation panel, select `LangGraph Cloud`. The `LangGraph Cloud` view contains a list of existing LangGraph Cloud deployments.
 1. Select an existing deployment to test with LangGraph Studio.
 1. In the top-right corner, select `Open LangGraph Studio`.
 1. [Invoke an assistant](./invoke_studio.md) or [view an existing thread](./threads_studio.md).

docs/docs/concepts/agentic_concepts.md

@@ -103,15 +103,15 @@ Parallel processing is vital for efficient multi-agent systems and complex tasks
 
 For practical implementation, see our [map-reduce tutorial](../how-tos/map-reduce.ipynb).
 
-### Sub-graphs
+### Subgraphs
 
-Sub-graphs are essential for managing complex agent architectures, particularly in multi-agent systems. They allow:
+[Subgraphs](./low_level.md#subgraphs) are essential for managing complex agent architectures, particularly in [multi-agent systems](./multi_agent.md). They allow:
 
 - Isolated state management for individual agents
 - Hierarchical organization of agent teams
 - Controlled communication between agents and the main system
 
-Sub-graphs communicate with the parent graph through overlapping keys in the state schema. This enables flexible, modular agent design. For implementation details, refer to our [sub-graph tutorial](../how-tos/subgraph.ipynb).
+Subgraphs communicate with the parent graph through overlapping keys in the state schema. This enables flexible, modular agent design. For implementation details, refer to our [subgraph how-to guide](../how-tos/subgraph.ipynb).
 
 ### Reflection
 

docs/docs/concepts/index.md

@@ -0,0 +1,27 @@
+---
+hide:
+  - navigation
+title: Concepts
+description: Conceptual Guide for LangGraph
+---
+
+# Conceptual Guide
+
+This guide provides explanations of the key concepts behind the LangGraph framework and AI applications more broadly.
+
+We recommend that you go through at least the [Quick Start](../tutorials/introduction.ipynb) before diving into the conceptual guide. This will provide practical context that will make it easier to understand the concepts discussed here.
+
+The conceptual guide does not cover step-by-step instructions or specific implementation examples — those are found in the [Tutorials](../tutorials/index.md) and [How-to guides](../how-tos/index.md). 
+For detailed reference material, please see the [API reference](../reference/index.md).
+
+## Concepts
+
+- [Why LangGraph?](high_level.md): A high-level overview of LangGraph and its goals.
+- [LangGraph Glossary](low_level.md): LangGraph workflows are designed as graphs, with nodes representing different components and edges representing the flow of information between them. This guide provides an overview of the key concepts associated with LangGraph graph primitives.
+- [Common Agentic Patterns](agentic_concepts.md): An agent are LLMs that can pick its own control flow to solve more complex problems! Agents are a key building block in many LLM applications. This guide explains the different types of agent architectures and how they can be used to control the flow of an application.
+- [Multi-Agent Systems](multi_agent.md): Complex LLM applications can often be broken down into multiple agents, each responsible for a different part of the application. This guide explains common patterns for building multi-agent systems.
+- [Human-in-the-Loop](human_in_the_loop.md): Explains different ways of integrating human feedback into a LangGraph application.
+- [Persistence](persistence.md): LangGraph has a built-in persistence layer, implemented through checkpointers. This persistence layer helps to support powerful capabilities like human-in-the-loop, memory, time travel, and fault-tolerance.
+- [Memory](memory.md): Memory in AI applications refers to the ability to process, store, and effectively recall information from past interactions. With memory, your agents can learn from feedback and adapt to users' preferences.  
+- [Streaming](streaming.md): Streaming is crucial for enhancing the responsiveness of applications built on LLMs. By displaying output progressively, even before a complete response is ready, streaming significantly improves user experience (UX), particularly when dealing with the latency of LLMs. 
+- [FAQ](faq.md): Frequently asked questions about LangGraph.

docs/docs/concepts/low_level.md

@@ -20,7 +20,7 @@ A super-step can be considered a single iteration over the graph nodes. Nodes th
 
 ### StateGraph
 
-The `StateGraph` class is the main graph class to uses. This is parameterized by a user defined `State` object.
+The `StateGraph` class is the main graph class to use. This is parameterized by a user defined `State` object.
 
 ### MessageGraph
 
@@ -52,12 +52,12 @@ By default, the graph will have the same input and output schemas. If you want t
 
 Typically, all graph nodes communicate with a single schema. This means that they will read and write to the same state channels. But, there are cases where we want more control over this:
 
-* Internal nodes can pass information that is not required in the graph's input / output.
-* We may also want to use different input / output schemas for the graph. The output might, for example, only contain a single relevant output key.
+- Internal nodes can pass information that is not required in the graph's input / output.
+- We may also want to use different input / output schemas for the graph. The output might, for example, only contain a single relevant output key.
 
-It is possible to have nodes write to private state channels inside the graph for internal node communication. We can simply define a private schema, `PrivateState`. See [this notebook](../how-tos/pass_private_state.ipynb) for more detail.  
+It is possible to have nodes write to private state channels inside the graph for internal node communication. We can simply define a private schema, `PrivateState`. See [this notebook](../how-tos/pass_private_state.ipynb) for more detail.
 
-It is also possible to define explicit input and output schemas for a graph. In these cases, we define an "internal" schema that contains *all* keys relevant to graph operations. But, we also define `input` and `output` schemas that are sub-sets of the "internal" schema to constrain the input and output of the graph. See [this notebook](../how-tos/input_output_schema.ipynb) for more detail.
+It is also possible to define explicit input and output schemas for a graph. In these cases, we define an "internal" schema that contains _all_ keys relevant to graph operations. But, we also define `input` and `output` schemas that are sub-sets of the "internal" schema to constrain the input and output of the graph. See [this notebook](../how-tos/input_output_schema.ipynb) for more detail.
 
 Let's look at an example:
 
@@ -101,11 +101,12 @@ graph = builder.compile()
 graph.invoke({"user_input":"My"})
 {'graph_output': 'My name is Lance'}
 ```
+
 There are two subtle and important points to note here:
 
-1. We pass `state: InputState` as the input schema to `node_1`. But, we write out to `foo`, a channel in `OverallState`. How can we write out to a state channel that is not included in the input schema? This is because a node *can write to any state channel in the graph state.* The graph state is the union of of the state channels defined at initialization, which includes `OverallState` and the filters `InputState` and `OutputState`.
+1. We pass `state: InputState` as the input schema to `node_1`. But, we write out to `foo`, a channel in `OverallState`. How can we write out to a state channel that is not included in the input schema? This is because a node _can write to any state channel in the graph state._ The graph state is the union of of the state channels defined at initialization, which includes `OverallState` and the filters `InputState` and `OutputState`.
 
-2. We initialize the graph with `StateGraph(OverallState,input=InputState,output=OutputState)`. So, how can we write to `PrivateState` in `node_2`? How does the graph gain access to this schema if it was not passed in the `StateGraph` initialization? We can do this because *nodes can also declare additional state channels* as long as the state schema definition exists. In this case, the `PrivateState` schema is defined, so we can add `bar` as a new state channel in the graph and write to it.
+2. We initialize the graph with `StateGraph(OverallState,input=InputState,output=OutputState)`. So, how can we write to `PrivateState` in `node_2`? How does the graph gain access to this schema if it was not passed in the `StateGraph` initialization? We can do this because _nodes can also declare additional state channels_ as long as the state schema definition exists. In this case, the `PrivateState` schema is defined, so we can add `bar` as a new state channel in the graph and write to it.
 
 ### Reducers
 
@@ -323,7 +324,16 @@ graph.add_conditional_edges("node_a", continue_to_jokes)
 
 ## Persistence
 
-LangGraph has a built-in persistence layer, implemented through [checkpointers][langgraph.checkpoint.base.BaseCheckpointSaver]. When you use a checkpointer with a graph, you can interact with and manage the graph's state after the execution. The checkpointer saves a _checkpoint_ (a snapshot) of the graph state at every superstep, enabling several powerful capabilities, including human-in-the-loop, memory and fault-tolerance. See this [conceptual guide](./persistence.md) for more information.
+LangGraph provides built-in persistence for your agent's state using [checkpointers][langgraph.checkpoint.base.BaseCheckpointSaver]. Checkpointers save snapshots of the graph state at every superstep, allowing resumption at any time. This enables features like human-in-the-loop interactions, memory management, and fault-tolerance. You can even directly manipulate a graph's state after its execution using the
+appropriate `get` and `update` methods. For more details, see the [persistence conceptual guide](./persistence.md).
+
+## Threads
+
+Threads in LangGraph represent individual sessions or conversations between your graph and a user. When using checkpointing, turns in a single conversation (and even steps within a single graph execution) are organized by a unique thread ID.
+
+## Storage
+
+LangGraph provides built-in document storage through the [BaseStore][langgraph.store.base.BaseStore] interface. Unlike checkpointers, which save state by thread ID, stores use custom namespaces for organizing data. This enables cross-thread persistence, allowing agents to maintain long-term memories, learn from past interactions, and accumulate knowledge over time. Common use cases include storing user profiles, building knowledge bases, and managing global preferences across all threads.
 
 ## Graph Migrations
 
@@ -407,10 +417,112 @@ def my_node(state: State) -> State:
     return state
 ```
 
+## Subgraphs
+
+A subgraph is a [graph](#graphs) that is used as a [node](#nodes) in another graph. This is nothing more than the age-old concept of encapsulation, applied to LangGraph. Some reasons for using subgraphs are:
+
+- building [multi-agent systems](./multi_agent.md)
+
+- when you want to reuse a set of nodes in multiple graphs, which maybe share some state, you can define them once in a subgraph and then use them in multiple parent graphs
+
+- when you want different teams to work on different parts of the graph independently, you can define each part as a subgraph, and as long as the subgraph interface (the input and output schemas) is respected, the parent graph can be built without knowing any details of the subgraph
+
+There are two ways to add subgraphs to a parent graph:
+
+- add a node with the compiled subgraph: this is useful when the parent graph and the subgraph share state keys and you don't need to transform state on the way in or out
+
+```python
+builder.add_node("subgraph", subgraph_builder.compile())
+```
+
+- add a node with a function that invokes the subgraph: this is useful when the parent graph and the subgraph have different state schemas and you need to transform state before or after calling the subgraph
+
+```python
+subgraph = subgraph_builder.compile()
+
+def call_subgraph(state: State):
+    return subgraph.invoke({"subgraph_key": state["parent_key"]})
+
+builder.add_node("subgraph", call_subgraph)
+```
+
+Let's take a look at examples for each.
+
+### As a compiled graph
+
+The simplest way to create subgraph nodes is by using a [compiled subgraph](#compiling-your-graph) directly. When doing so, it is **important** that the parent graph and the subgraph [state schemas](#state) share at least one key which they can use to communicate. If your graph and subgraph do not share any keys, you should use write a function [invoking the subgraph](#as-a-function) instead.
+
+!!! Note
+    If you pass extra keys to the subgraph node (i.e., in addition to the shared keys), they will be ignored by the subgraph node. Similarly, if you return extra keys from the subgraph, they will be ignored by the parent graph.
+
+```python
+from langgraph.graph import START, StateGraph
+from typing import TypedDict
+
+class State(TypedDict):
+    foo: str
+
+class SubgraphState(TypedDict):
+    foo: str  # note that this key is shared with the parent graph state
+    bar: str
+
+# Define subgraph
+def subgraph_node(state: SubgraphState):
+    # note that this subgraph node can communicate with the parent graph via the shared "foo" key
+    return {"foo": state["foo"] + "bar"}
+
+subgraph_builder = StateGraph(SubgraphState)
+subgraph_builder.add_node(subgraph_node)
+...
+subgraph = subgraph_builder.compile()
+
+# Define parent graph
+builder = StateGraph(State)
+builder.add_node("subgraph", subgraph)
+...
+graph = builder.compile()
+```
+
+### As a function
+
+You might want to define a subgraph with a completely different schema. In this case, you can create a node function that invokes the subgraph. This function will need to [transform](../how-tos/subgraph-transform-state.ipynb) the input (parent) state to the subgraph state before invoking the subgraph, and transform the results back to the parent state before returning the state update from the node.
+
+```python
+class State(TypedDict):
+    foo: str
+
+class SubgraphState(TypedDict):
+    # note that none of these keys are shared with the parent graph state
+    bar: str
+    baz: str
+
+# Define subgraph
+def subgraph_node(state: SubgraphState):
+    return {"bar": state["bar"] + "baz"}
+
+subgraph_builder = StateGraph(SubgraphState)
+subgraph_builder.add_node(subgraph_node)
+...
+subgraph = subgraph_builder.compile()
+
+# Define parent graph
+def node(state: State):
+    # transform the state to the subgraph state
+    response = subgraph.invoke({"bar": state["foo"]})
+    # transform response back to the parent state
+    return {"foo": response["bar"]}
+
+builder = StateGraph(State)
+# note that we are using `node` function instead of a compiled subgraph
+builder.add_node(node)
+...
+graph = builder.compile()
+```
+
 ## Visualization
 
 It's often nice to be able to visualize graphs, especially as they get more complex. LangGraph comes with several built-in ways to visualize graphs. See [this how-to guide](../how-tos/visualization.ipynb) for more info.
 
 ## Streaming
 
-LangGraph is built with first class support for streaming, including streaming updates from graph nodes during the execution, streaming tokens from LLM calls and more. See this [conceptual guide](./streaming.md) for more information.
+LangGraph is built with first class support for streaming, including streaming updates from graph nodes during the execution, streaming tokens from LLM calls and more. See this [conceptual guide](./streaming.md) for more information.

docs/docs/concepts/memory.md

@@ -2,41 +2,145 @@
 
 ## What is Memory?
 
-Memory in the context of LLMs and AI applications refers to the ability to process, retain, and utilize information from past interactions or data sources. Examples include:
+[Memory](https://pmc.ncbi.nlm.nih.gov/articles/PMC10410470/) is a cognitive function that allows people to store, retrieve, and use information to understand their present and future. Consider the frustration of working with a colleague who forgets everything you tell them, requiring constant repetition! As AI agents undertake more complex tasks involving numerous user interactions, equipping them with memory becomes equally crucial for efficiency and user satisfaction. With memory, agents can learn from feedback and adapt to users' preferences. This guide covers two types of memory based on recall scope:
 
-- Managing what messages (e.g., from a long message history) are sent to a chat model to limit token usage
-- Summarizing past conversations to give a chat model context from prior interactions
-- Selecting few shot examples (e.g., from a dataset) to guide model responses
-- Maintaining persistent data (e.g., user preferences) across multiple chat sessions
-- Allowing an LLM to update its own prompt using past information (e.g., meta-prompting)
-- Retrieving information relevant to a conversation or question from a long-term storage system
+**Short-term memory**, or [thread](persistence.md#threads)-scoped memory, can be recalled at any time **from within** a single conversational thread with a user. LangGraph manages short-term memory as a part of your agent's [state](low_level.md#state). State is persisted to a database using a [checkpointer](persistence.md#checkpoints) so the thread can be resumed at any time. Short-term memory updates when the graph is invoked or a step is completed, and the State is read at the start of each step.
 
-Below, we'll discuss each of these examples in some detail. 
+**Long-term memory** is shared **across** conversational threads. It can be recalled _at any time_ and **in any thread**. Memories are scoped to any custom namespace, not just within a single thread ID. LangGraph provides [stores](persistence.md#memory-store) ([reference doc](https://langchain-ai.github.io/langgraph/reference/store/#langgraph.store.base.BaseStore)) to let you save and recall long-term memories.
 
-## Managing Messages
+Both are important to understand and implement for your application.
+
+![](img/memory/short-vs-long.png)
+
+## Short-term memory
+
+Short-term memory lets your application remember previous interactions within a single [thread](persistence.md#threads) or conversation. A [thread](persistence.md#threads) organizes multiple interactions in a session, similar to the way email groups messages in a single conversation.
+
+LangGraph manages short-term memory as part of the agent's state, persisted via thread-scoped checkpoints. This state can normally include the conversation history along with other stateful data, such as uploaded files, retrieved documents, or generated artifacts. By storing these in the graph's state, the bot can access the full context for a given conversation while maintaining separation between different threads.
+
+Since conversation history is the most common form of representing short-term memory, in the next section, we will cover techniques for managing conversation history when the list of messages becomes **long**. If you want to stick to the high-level concepts, continue on to the [long-term memory](#long-term-memory) section.
+
+### Managing long conversation history
+
+Long conversations pose a challenge to today's LLMs. The full history may not even fit inside an LLM's context window, resulting in an irrecoverable error. Even _if_ your LLM technically supports the full context length, most LLMs still perform poorly over long contexts. They get "distracted" by stale or off-topic content, all while suffering from slower response times and higher costs.
+
+Managing short-term memory is an exercise of balancing [precision & recall](https://en.wikipedia.org/wiki/Precision_and_recall#:~:text=Precision%20can%20be%20seen%20as,irrelevant%20ones%20are%20also%20returned) with your application's other performance requirements (latency & cost). As always, it's important to think critically about how you represent information for your LLM and to look at your data. We cover a few common techniques for managing message lists below and hope to provide sufficient context for you to pick the best tradeoffs for your application:
+
+- [Editing message lists](#editing-message-lists): How to think about trimming and filtering a list of messages before passing to language model.
+- [Summarizing past conversations](#summarizing-past-conversations): A common technique to use when you don't just want to filter the list of messages.
 
 ### Editing message lists
 
-Chat models accept instructions through [messages](https://python.langchain.com/docs/concepts/#messages), which can serve as general instructions (e.g., a system message) or user-provided instructions (e.g., human messages). In chat applications, messages often alternate between human inputs and model responses, accumulating in a list over time. Because context windows are limited and token-rich message lists can be costly, many applications can benefit from approaches to actively manage messages.    
+Chat models accept context using [messages](https://python.langchain.com/docs/concepts/#messages), which include developer provided instructions (a system message) and user inputs (human messages). In chat applications, messages alternate between human inputs and model responses, resulting in a list of messages that grows longer over time. Because context windows are limited and token-rich message lists can be costly, many applications can benefit from using techniques to manually remove or forget stale information.
 
-The most directed approach is to remove specific messages from a list. This can be done using [RemoveMessage](https://langchain-ai.github.io/langgraph/how-tos/memory/delete-messages/#manually-deleting-messages) based upon the message `id`, a unique identifier for each message. In the below example, we keep only the last two messages in the list using `RemoveMessage` to remove older messages based  upon their `id`.
+![](img/memory/filter.png)
+
+The most direct approach is to remove old messages from a list (similar to a [least-recently used cache](https://en.wikipedia.org/wiki/Page_replacement_algorithm#Least_recently_used)).
+
+The typical technique for deleting content from a list in LangGraph is to return an update from a node telling the system to delete some portion of the list. You get to define what this update looks like, but a common approach would be to let you return an object or dictionary specifying which values to retain.
 
 ```python
-from langchain_core.messages import RemoveMessage
+def manage_list(existing: list, updates: Union[list, dict]):
+    if isinstance(updates, list):
+        # Normal case, add to the history
+        return existing + updates
+    elif isinstance(updates, dict) and updates["type"] == "keep":
+        # You get to decide what this looks like.
+        # For example, you could simplify and just accept a string "DELETE"
+        # and clear the entire list.
+        return existing[updates["from"]:updates["to"]]
+    # etc. We define how to interpret updates
 
-# Message list
-messages = [AIMessage("Hi.", name="Bot", id="1")]
-messages.append(HumanMessage("Hi.", name="Lance", id="2"))
-messages.append(AIMessage("So you said you were researching ocean mammals?", name="Bot", id="3"))
-messages.append(HumanMessage("Yes, I know about whales. But what others should I learn about?", name="Lance", id="4"))
+class State(TypedDict):
+    my_list: Annotated[list, manage_list]
 
-# Isolate messages to delete
-delete_messages = [RemoveMessage(id=m.id) for m in messages[:-2]]
-print(delete_messages)
-[RemoveMessage(content='', id='1'), RemoveMessage(content='', id='2')]
+def my_node(state: State):
+    return {
+        # We return an update for the field "my_list" saying to
+        # keep only values from index -5 to the end (deleting the rest)
+        "my_list": {"type": "keep", "from": -5, "to": None}
+    }
 ```
 
-Because the context window for chat model is denominated in tokens, it can be useful to trim message lists based upon some number of tokens that we want to retain. To do this, we can use [`trim_messages`](https://python.langchain.com/docs/how_to/trim_messages/#trimming-based-on-token-count) and specify number of token to keep from the list, as well as the `strategy` (e.g., keep the last `max_tokens`). 
+LangGraph will call the `manage_list` "[reducer](low_level.md#reducers)" function any time an update is returned under the key "my_list". Within that function, we define what types of updates to accept. Typically, messages will be added to the existing list (the conversation will grow); however, we've also added support to accept a dictionary that lets you "keep" certain parts of the state. This lets you programmatically drop old message context.
+
+Another common approach is to let you return a list of "remove" objects that specify the IDs of all messages to delete. If you're using the LangChain messages and the [`add_messages`](https://langchain-ai.github.io/langgraph/reference/graphs/#langgraph.graph.message.add_messages) reducer (or `MessagesState`, which uses the same underlying functionality) in LangGraph, you can do this using a `RemoveMessage`.
+
+```python
+from langchain_core.messages import RemoveMessage, AIMessage
+from langgraph.graph import add_messages
+# ... other imports
+
+class State(TypedDict):
+    # add_messages will default to upserting messages by ID to the existing list
+    # if a RemoveMessage is returned, it will delete the message in the list by ID
+    messages: Annotated[list, add_messages]
+
+def my_node_1(state: State):
+    # Add an AI message to the `messages` list in the state
+    return {"messages": [AIMessage(content="Hi")]}
+
+def my_node_2(state: State):
+    # Delete all but the last 2 messages from the `messages` list in the state
+    delete_messages = [RemoveMessage(id=m.id) for m in state['messages'][:-2]]
+    return {"messages": delete_messages}
+
+```
+
+In the example above, the `add_messages` reducer allows us to [append](https://langchain-ai.github.io/langgraph/concepts/low_level/#serialization) new messages to the `messages` state key as shown in `my_node_1`. When it sees a `RemoveMessage`, it will delete the message with that ID from the list (and the RemoveMessage will then be discarded). For more information on LangChain-specific message handling, check out [this how-to on using `RemoveMessage` ](https://langchain-ai.github.io/langgraph/how-tos/memory/delete-messages/).
+
+See this how-to [guide](https://langchain-ai.github.io/langgraph/how-tos/memory/manage-conversation-history/) and module 2 from our [LangChain Academy](https://github.com/langchain-ai/langchain-academy/tree/main/module-2) course for example usage.
+
+### Summarizing past conversations
+
+The problem with trimming or removing messages, as shown above, is that we may lose information from culling of the message queue. Because of this, some applications benefit from a more sophisticated approach of summarizing the message history using a chat model.
+
+![](img/memory/summary.png)
+
+Simple prompting and orchestration logic can be used to achieve this. As an example, in LangGraph we can extend the [MessagesState](https://langchain-ai.github.io/langgraph/concepts/low_level/#working-with-messages-in-graph-state) to include a `summary` key.
+
+```python
+from langgraph.graph import MessagesState
+class State(MessagesState):
+    summary: str
+```
+
+Then, we can generate a summary of the chat history, using any existing summary as context for the next summary. This `summarize_conversation` node can be called after some number of messages have accumulated in the `messages` state key.
+
+```python
+def summarize_conversation(state: State):
+
+    # First, we get any existing summary
+    summary = state.get("summary", "")
+
+    # Create our summarization prompt
+    if summary:
+
+        # A summary already exists
+        summary_message = (
+            f"This is a summary of the conversation to date: {summary}\n\n"
+            "Extend the summary by taking into account the new messages above:"
+        )
+
+    else:
+        summary_message = "Create a summary of the conversation above:"
+
+    # Add prompt to our history
+    messages = state["messages"] + [HumanMessage(content=summary_message)]
+    response = model.invoke(messages)
+
+    # Delete all but the 2 most recent messages
+    delete_messages = [RemoveMessage(id=m.id) for m in state["messages"][:-2]]
+    return {"summary": response.content, "messages": delete_messages}
+```
+
+See this how-to [here](https://langchain-ai.github.io/langgraph/how-tos/memory/add-summary-conversation-history/) and module 2 from our [LangChain Academy](https://github.com/langchain-ai/langchain-academy/tree/main/module-2) course for example usage.
+
+### Knowing **when** to remove messages
+
+Most LLMs have a maximum supported context window (denominated in tokens). A simple way to decide when to truncate messages is to count the tokens in the message history and truncate whenever it approaches that limit. Naive truncation is straightforward to implement on your own, though there are a few "gotchas". Some model APIs further restrict the sequence of message types (must start with human message, cannot have consecutive messages of the same type, etc.). If you're using LangChain, you can use the [`trim_messages`](https://python.langchain.com/docs/how_to/trim_messages/#trimming-based-on-token-count) utility and specify the number of tokens to keep from the list, as well as the `strategy` (e.g., keep the last `max_tokens`) to use for handling the boundary.
+
+Below is an example.
 
 ```python
 from langchain_core.messages import trim_messages
@@ -46,10 +150,7 @@ trim_messages(
     strategy="last",
     # Remember to adjust based on your model
     # or else pass a custom token_encoder
-    token_counter=ChatOpenAI(model="gpt-4o"),
-    # Most chat models expect that chat history starts with either:
-    # (1) a HumanMessage or
-    # (2) a SystemMessage followed by a HumanMessage
+    token_counter=ChatOpenAI(model="gpt-4"),
     # Remember to adjust based on the desired conversation
     # length
     max_tokens=45,
@@ -67,109 +168,141 @@ trim_messages(
     include_system=True,
 )
 ```
-### Usage with LangGraph
 
-When building agents in LangGraph, we commonly want to manage a list of messages in the graph state. Because this is such a common use case, [MessagesState](https://langchain-ai.github.io/langgraph/concepts/low_level/#working-with-messages-in-graph-state) is a built-in LangGraph state schema that includes a `messages` key, which is a list of messages. `MessagesState` also includes an `add_messages` reducer for updating the messages list with new messages as the application runs. The `add_messages` reducer allows us to [append](https://langchain-ai.github.io/langgraph/concepts/low_level/#serialization) new messages to the `messages` state key as shown below. When we perform a state update with `{"messages": new_message}` returned from `my_node`, the `add_messages` reducer appends `new_message` to the existing list of messages.
+## Long-term memory
+
+Long-term memory in LangGraph allows systems to retain information across different conversations or sessions. Unlike short-term memory, which is thread-scoped, long-term memory is saved within custom "namespaces."
+
+### Storing memories
+
+LangGraph stores long-term memories as JSON documents in a [store](persistence.md#memory-store) ([reference doc](https://langchain-ai.github.io/langgraph/reference/store/#langgraph.store.base.BaseStore)). Each memory is organized under a custom `namespace` (similar to a folder) and a distinct `key` (like a filename). Namespaces often include user or org IDs or other labels that makes it easier to organize information. This structure enables hierarchical organization of memories. Cross-namespace searching is then supported through content filters. See the example below for an example.
 
 ```python
-def my_node(state: State):
-    # Add a new message to the state
-    new_message = HumanMessage(content="message")
-    return {"messages": new_message}
+from langgraph.store.memory import InMemoryStore
+
+# InMemoryStore saves data to an in-memory dictionary. Use a DB-backed store in production use.
+store = InMemoryStore()
+user_id = "my-user"
+application_context = "chitchat"
+namespace = (user_id, application_context)
+store.put(namespace, "a-memory", {"rules": ["User likes short, direct language", "User only speaks English & python"], "my-key": "my-value"})
+# get the "memory" by ID
+item = store.get(namespace, "a-memory")
+# list "memories" within this namespace, filtering on content equivalence
+items = store.search(namespace, filter={"my-key": "my-value"})
 ```
 
-The `add_messages` reducer built into `MessagesState` [also works with the `RemoveMessage` utility that we discussed above](https://langchain-ai.github.io/langgraph/how-tos/memory/delete-messages/). In this case, we can perform a state update with a list of `delete_messages` to remove specific messages from the `messages` list.
+### Framework for thinking about long-term memory
 
-```python
-def my_node(state: State):
-    # Delete messages from state
-    delete_messages = [RemoveMessage(id=m.id) for m in state['messages'][:-2]]
-    return {"messages": delete_messages}
-```
+Long-term memory is a complex challenge without a one-size-fits-all solution. However, the following questions provide a structure framework to help you navigate the different techniques:
 
-See this how-to [guide](https://langchain-ai.github.io/langgraph/how-tos/memory/manage-conversation-history/) and module 2 from our [LangChain Academy](https://github.com/langchain-ai/langchain-academy/tree/main/module-2) course for example usage.
+**What is the type of memory?**
 
-## Summarizing Past Conversations
+Humans use memories to remember [facts](https://en.wikipedia.org/wiki/Semantic_memory), [experiences](https://en.wikipedia.org/wiki/Episodic_memory), and [rules](https://en.wikipedia.org/wiki/Procedural_memory). AI agents can use memory in the same ways. For example, AI agents can use memory to remember specific facts about a user to accomplish a task. We expand on several types of memories in the [section below](#memory-types).
 
-The problem with trimming or removing messages, as shown above, is that we may loose information from culling of the message queue. Because of this, some applications benefit from a more sophisticated approach of summarizing the message history using a chat model. 
+**When do you want to update memories?**
 
-Simple prompting and orchestration logic can be used to achieve this. As an example, in LangGraph we can extend the [MessagesState](https://langchain-ai.github.io/langgraph/concepts/low_level/#working-with-messages-in-graph-state) to include a `summary` key. 
+Memory can be updated as part of an agent's application logic (e.g. "on the hot path"). In this case, the agent typically decides to remember facts before responding to a user. Alternatively, memory can be updated as a background task (logic that runs in the background / asynchronously and generates memories). We explain the tradeoffs between these approaches in the [section below](#writing-memories).
 
-```python
-from langgraph.graph import MessagesState
-class State(MessagesState):
-    summary: str
-```
+## Memory types
 
-Then, we can generate a summary of the chat history, using any existing summary as context for the next summary. This `summarize_conversation` node can be called after some number of messages have accumulated in the `messages` state key.
+Different applications require various types of memory. Although the analogy isn't perfect, examining [human memory types](https://www.psychologytoday.com/us/basics/memory/types-of-memory?ref=blog.langchain.dev) can be insightful. Some research (e.g., the [CoALA paper](https://arxiv.org/pdf/2309.02427)) have even mapped these human memory types to those used in AI agents.
 
-```python   
-def summarize_conversation(state: State):
-    
-    # First, we get any existing summary
-    summary = state.get("summary", "")
+| Memory Type | What is Stored | Human Example | Agent Example |
+|-------------|----------------|---------------|---------------|
+| Semantic | Facts | Things I learned in school | Facts about a user |
+| Episodic | Experiences | Things I did | Past agent actions |
+| Procedural | Instructions | Instincts or motor skills | Agent system prompt |
 
-    # Create our summarization prompt 
-    if summary:
-        
-        # A summary already exists
-        summary_message = (
-            f"This is summary of the conversation to date: {summary}\n\n"
-            "Extend the summary by taking into account the new messages above:"
-        )
-        
-    else:
-        summary_message = "Create a summary of the conversation above:"
+### Semantic Memory
 
-    # Add prompt to our history
-    messages = state["messages"] + [HumanMessage(content=summary_message)]
-    response = model.invoke(messages)
-    
-    # Delete all but the 2 most recent messages
-    delete_messages = [RemoveMessage(id=m.id) for m in state["messages"][:-2]]
-    return {"summary": response.content, "messages": delete_messages}
-```
+[Semantic memory](https://en.wikipedia.org/wiki/Semantic_memory), both in humans and AI agents, involves the retention of specific facts and concepts. In humans, it can include information learned in school and the understanding of concepts and their relationships. For AI agents, semantic memory is often used to personalize applications by remembering facts or concepts from past interactions. 
 
-See this how-to [here](https://langchain-ai.github.io/langgraph/how-tos/memory/add-summary-conversation-history/) and module 2 from our [LangChain Academy](https://github.com/langchain-ai/langchain-academy/tree/main/module-2) course for example usage.
+#### Profile
 
-## Few Shot Examples
+Semantic memories can be managed in different ways. For example, memories can be a single, continuously updated "profile" of well-scoped and specific information about a user, organization, or other entity (including the agent itself). A profile is generally just a JSON document with various key-value pairs you've selected to represent your domain. 
 
-Few-shot learning is a powerful technique where LLMs can be ["programmed"](https://x.com/karpathy/status/1627366413840322562) inside the prompt with input-output examples to perform diverse tasks. While various [best-practices](https://python.langchain.com/docs/concepts/#1-generating-examples) can be used to generate few-shot examples, often the challenge lies in selecting the most relevant examples based on user input. 
+When remembering a profile, you will want to make sure that you are **updating** the profile each time. As a result, you will want to pass in the previous profile and [ask the model to generate a new profile](https://github.com/langchain-ai/memory-template) (or some [JSON patch](https://github.com/hinthornw/trustcall) to apply to the old profile). This can be become error-prone as the profile gets larger, and may benefit from splitting a profile into multiple documents or **strict** decoding when generating documents to ensure the memory schemas remains valid.
 
-LangChain [`ExampleSelectors`](https://python.langchain.com/docs/how_to/#example-selectors) can be used to customize few-shot example selection from a collection of examples using criteria such as length, semantic similarity, semantic ngram overlap, or maximal marginal relevance.
+![](img/memory/update-profile.png)
 
-If few-shot examples are stored in a [LangSmith Dataset](https://docs.smith.langchain.com/how_to_guides/datasets), then dynamic few-shot example selectors can be used out-of-the box to achieve this same goal. LangSmith will index the dataset for you and enable retrieval of few shot examples that are most relevant to the user input based upon keyword similarity ([using a BM25-like algorithm](https://docs.smith.langchain.com/how_to_guides/datasets/index_datasets_for_dynamic_few_shot_example_selection) for keyword based similarity). 
+#### Collection
+
+Alternatively, memories can be a collection of documents that are continuously updated and extended over time. Each individual memory can be more narrowly scoped and easier to generate, which means that you're less likely to **lose** information over time. It's easier for an LLM to generate _new_ objects for new information than reconcile new information with an existing profile. As a result, a document collection tends to lead to [higher recall downstream](https://en.wikipedia.org/wiki/Precision_and_recall).
+
+However, this shifts some complexity memory updating. The model must now _delete_ or _update_ existing items in the list, which can be tricky. In addition, some models may default to over-inserting and others may default to over-updating. See the [Trustcall](https://github.com/hinthornw/trustcall) package for one way to manage this and consider evaluation (e.g., with a tool like [LangSmith](https://docs.smith.langchain.com/tutorials/Developers/evaluation)) to help you tune the behavior.
+
+Working with document collections also shifts complexity to memory **search** over the list. The `Store` currently supports [filtering by metadata](https://langchain-ai.github.io/langgraph/reference/store/#storage) and will soon add [semantic search shortly](https://python.langchain.com/docs/concepts/vectorstores/), but selecting the most relevant documents can be tricky as the list grows.
+
+Finally, using a collection of memories can make it challenging to provide comprehensive context to the model. While individual memories may follow a specific schema, this structure might not capture the full context or relationships between memories. As a result, when using these memories to generate responses, the model may lack important contextual information that would be more readily available in a unified profile approach.
+
+![](img/memory/update-list.png)
+
+Regardless of memory management approach, the central point is that the agent will use the semantic memories to [ground its responses](https://python.langchain.com/docs/concepts/rag/), which often leads to more personalized and relevant interactions.
+
+### Episodic Memory
+
+[Episodic memory](https://en.wikipedia.org/wiki/Episodic_memory), in both humans and AI agents, involves recalling past events or actions. The [CoALA paper](https://arxiv.org/pdf/2309.02427) frames this well: facts can be written to semantic memory, whereas *experiences* can be written to episodic memory. For AI agents, episodic memory is often used to help an agent remember how to accomplish a task. 
+
+In practice, episodic memories are often implemented through [few-shot example prompting](https://python.langchain.com/docs/concepts/few_shot_prompting/), where agents learn from past sequences to perform tasks correctly. Sometimes it's easier to "show" than "tell" and LLMs learn well from examples. Few-shot learning lets you ["program"](https://x.com/karpathy/status/1627366413840322562) your LLM by updating the prompt with input-output examples to illustrate the intended behavior. While various [best-practices](https://python.langchain.com/docs/concepts/#1-generating-examples) can be used to generate few-shot examples, often the challenge lies in selecting the most relevant examples based on user input.
+
+Note that the memory [store](persistence.md#memory-store) is just one way to store data as few-shot examples. If you want to have more developer involvement, or tie few-shots more closely to your evaluation harness, you can also use a [LangSmith Dataset](https://docs.smith.langchain.com/evaluation/how_to_guides/datasets/index_datasets_for_dynamic_few_shot_example_selection) to store your data. Then dynamic few-shot example selectors can be used out-of-the box to achieve this same goal. LangSmith will index the dataset for you and enable retrieval of few shot examples that are most relevant to the user input based upon keyword similarity ([using a BM25-like algorithm](https://docs.smith.langchain.com/how_to_guides/datasets/index_datasets_for_dynamic_few_shot_example_selection) for keyword based similarity). 
 
 See this how-to [video](https://www.youtube.com/watch?v=37VaU7e7t5o) for example usage of dynamic few-shot example selection in LangSmith. Also, see this [blog post](https://blog.langchain.dev/few-shot-prompting-to-improve-tool-calling-performance/) showcasing few-shot prompting to improve tool calling performance and this [blog post](https://blog.langchain.dev/aligning-llm-as-a-judge-with-human-preferences/) using few-shot example to align an LLMs to human preferences.
 
-## Maintaining Data Across Chat Sessions
+### Procedural Memory
 
-LangGraph's [persistence layer](https://langchain-ai.github.io/langgraph/concepts/persistence/#persistence) has checkpointers that utilize various storage systems, including an in-memory key-value store or different databases. These checkpoints capture the graph state at each execution step and accumulate in a thread, which can be accessed at a later time using a thread ID to resume a previous graph execution. We add persistence to our graph by passing a checkpointer to the `compile` method, as shown here.
+[Procedural memory](https://en.wikipedia.org/wiki/Procedural_memory), in both humans and AI agents, involves remembering the rules used to perform tasks. In humans, procedural memory is like the internalized knowledge of how to perform tasks, such as riding a bike via basic motor skills and balance. Episodic memory, on the other hand, involves recalling specific experiences, such as the first time you successfully rode a bike without training wheels or a memorable bike ride through a scenic route. For AI agents, procedural memory is a combination of model weights, agent code, and agent's prompt that collectively determine the agent's functionality. 
+
+In practice, it is fairly uncommon for agents to modify their model weights or rewrite their code. However, it is more common for agents to [modify their own prompts](https://docs.anthropic.com/en/docs/build-with-claude/prompt-engineering/prompt-generator). 
+
+One effective approach to refining an agent's instructions is through ["Reflection"](https://blog.langchain.dev/reflection-agents/) or meta-prompting. This involves prompting the agent with its current instructions (e.g., the system prompt) along with recent conversations or explicit user feedback. The agent then refines its own instructions based on this input. This method is particularly useful for tasks where instructions are challenging to specify upfront, as it allows the agent to learn and adapt from its interactions.
+
+For example, we built a [Tweet generator](https://www.youtube.com/watch?v=Vn8A3BxfplE) using external feedback and prompt re-writing to produce high-quality paper summaries for Twitter. In this case, the specific summarization prompt was difficult to specify *a priori*, but it was fairly easy for a user to critique the generated Tweets and provide feedback on how to improve the summarization process. 
+
+The below pseudo-code shows how you might implement this with the LangGraph memory [store](persistence.md#memory-store), using the store to save a prompt, the `update_instructions` node to get the current prompt (as well as feedback from the conversation with the user captured in `state["messages"]`), update the prompt, and save the new prompt back to the store. Then, the `call_model` get the updated prompt from the store and uses it to generate a response.
 
 ```python
-# Compile the graph with a checkpointer
-checkpointer = MemorySaver()
-graph = workflow.compile(checkpointer=checkpointer)
+# Node that *uses* the instructions
+def call_model(state: State, store: BaseStore):
+    namespace = ("agent_instructions", )
+    instructions = store.get(namespace, key="agent_a")[0]
+    # Application logic
+    prompt = prompt_template.format(instructions=instructions.value["instructions"])
+    ...
 
-# Invoke the graph with a thread ID
-config = {"configurable": {"thread_id": "1"}}
-graph.invoke(input_state, config)
-
-# get the latest state snapshot at a later time
-config = {"configurable": {"thread_id": "1"}}
-graph.get_state(config)
+# Node that updates instructions
+def update_instructions(state: State, store: BaseStore):
+    namespace = ("instructions",)
+    current_instructions = store.search(namespace)[0]
+    # Memory logic
+    prompt = prompt_template.format(instructions=instructions.value["instructions"], conversation=state["messages"])
+    output = llm.invoke(prompt)
+    new_instructions = output['new_instructions']
+    store.put(("agent_instructions",), "agent_a", {"instructions": new_instructions})
+    ...
 ```
 
-Persistence is critical sustaining a long-running chat sessions. For example, a chat between a user and an AI assistant may have interruptions. Persistence ensures that a user can continue that particular chat session at any later point in time. However, what happens if a user initiates a new chat session with an assistant? This spawns a new thread, and the information from the previous session (thread) is not retained. This motivates the need for a memory service that can maintain data across chat sessions (threads). 
+![](img/memory/update-instructions.png)
 
-## Meta-prompting
+## Writing memories
 
-Meta-prompting uses an LLM to generate or refine its own prompts or instructions. This approach allows the system to dynamically update and improve its own behavior, potentially leading to better performance on various tasks. This is particularly useful for tasks where the instructions are challenging to specify a priori. 
+While [humans often form long-term memories during sleep](https://medicine.yale.edu/news-article/sleeps-crucial-role-in-preserving-memory/), AI agents need a different approach. When and how should agents create new memories? There are at least two primary methods for agents to write memories: "on the hot path" and "in the background".
 
-Meta-prompting can use past information to update the prompt. As an example, this [Tweet generator](https://www.youtube.com/watch?v=Vn8A3BxfplE) uses meta-prompting to iteratively improve the summarization prompt used to generate high quality paper summaries for Twitter. In this case, we used a LangSmith dataset to house several papers that we wanted to summarize, generated summaries using a naive summarization prompt, manually reviewed the summaries, captured feedback from human review using the LangSmith Annotation Queue, and passed this feedback to a chat model to re-generate the summarization prompt. The process was repeated in a loop until the summaries met our criteria in human review.
+![](img/memory/hot_path_vs_background.png)
 
-## Retrieving relevant information from long-term storage
+### Writing memories in the hot path
 
-A central challenge that spans many different memory use-case can be summarized simply: how can we retrieve *relevant information* from a long-term storage system and pass it to a chat model? As an example, assume we have a system that stores a large number of specific details about a user, but the user asks a specific question related to restaurant recommendations. It would be costly to trivially extract *all* personal user information and pass it to a chat model. Instead, we want to extract only the information that is most relevant to the user's current chat interaction (e,g,. food preferences, location, etc.) and pass it to the chat model. 
+Creating memories during runtime offers both advantages and challenges. On the positive side, this approach allows for real-time updates, making new memories immediately available for use in subsequent interactions. It also enables transparency, as users can be notified when memories are created and stored.
 
-There is a large body of work on retrieval that aims to address this challenge. See our tutorials focused on [RAG, or Retrieval Augmented Generation](https://langchain-ai.github.io/langgraph/tutorials/rag/langgraph_adaptive_rag/), our conceptual docs on [retrieval](https://python.langchain.com/docs/concepts/#retrieval), and our [open source repository](https://github.com/langchain-ai/rag-from-scratch) along with [videos](https://www.youtube.com/playlist?list=PLfaIDFEXuae2LXbO1_PKyVJiQ23ZztA0x) on this topic.
+However, this method also presents challenges. It may increase complexity if the agent requires a new tool to decide what to commit to memory. In addition, the process of reasoning about what to save to memory can impact agent latency. Finally, the agent must multitask between memory creation and its other responsibilities, potentially affecting the quantity and quality of memories created.
+
+As an example, ChatGPT uses a [save_memories](https://openai.com/index/memory-and-new-controls-for-chatgpt/) tool to upsert memories as content strings, deciding whether and how to use this tool with each user message. See our [memory-agent](https://github.com/langchain-ai/memory-agent) template as an reference implementation.
+
+### Writing memories in the background
+
+Creating memories as a separate background task offers several advantages. It eliminates latency in the primary application, separates application logic from memory management, and allows for more focused task completion by the agent. This approach also provides flexibility in timing memory creation to avoid redundant work.
+
+However, this method has its own challenges. Determining the frequency of memory writing becomes crucial, as infrequent updates may leave other threads without new context. Deciding when to trigger memory formation is also important. Common strategies include scheduling after a set time period (with rescheduling if new events occur), using a cron schedule, or allowing manual triggers by users or the application logic.
+
+See our [memory-service](https://github.com/langchain-ai/memory-template) template as an reference implementation.

docs/docs/concepts/multi_agent.md

@@ -1,138 +1,281 @@
 # Multi-agent Systems
 
-A multi-agent system is a system with multiple independent actors powered by LLMs that are connected in a specific way. These actors can be as simple as a prompt and an LLM call, or as complex as a [ReAct](./agentic_concepts.md#react-implementation) agent.
+An [agent](./agentic_concepts.md#agent-architectures) is _a system that uses an LLM to decide the control flow of an application_. As you develop these systems, they might grow more complex over time, making them harder to manage and scale. For example, you might run into the following problems:
 
-The primary benefits of this architecture are:
+- agent has too many tools at its disposal and makes poor decisions about which tool to call next
+- context grows too complex for a single agent to keep track of
+- there is a need for multiple specialization areas in the system (e.g. planner, researcher, math expert, etc.)
 
-* **Modularity**: Separate agents facilitate easier development, testing, and maintenance of agentic systems.
-* **Specialization**: You can create expert agents focused on specific domains, and compose them into more complex applications
-* **Control**: You can explicitly control how agents communicate (as opposed to relying on function calling)
+To tackle these, you might consider breaking your application into multiple smaller, independent agents and composing them into a **multi-agent system**. These independent agents can be as simple as a prompt and an LLM call, or as complex as a [ReAct](./agentic_concepts.md#react-implementation) agent (and more!).
 
-## Multi-agent systems in LangGraph
+The primary benefits of using multi-agent systems are:
 
-### Agents as nodes
+- **Modularity**: Separate agents make it easier to develop, test, and maintain agentic systems.
+- **Specialization**: You can create expert agents focused on specific domains, which helps with the overall system performance.
+- **Control**: You can explicitly control how agents communicate (as opposed to relying on function calling).
 
-Agents can be defined as nodes in LangGraph. As any other node in the LangGraph, these agent nodes receive the graph state as an input and return an update to the state as their output.
+## Multi-agent architectures
 
-* Simple **LLM nodes**: single LLMs with custom prompts
-* **Subgraph nodes**: complex graphs called inside the orchestrator graph node
+![](./img/multi_agent/architectures.png)
 
-![](./img/multi_agent/subgraph.png)
+There are several ways to connect agents in a multi-agent system:
 
-### Agents as tools
+- **Network**: each agent can communicate with [every other agent](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/multi-agent-collaboration/). Any agent can decide which other agent to call next.
+- **Supervisor**: each agent communicates with a single [supervisor](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/agent_supervisor/) agent. Supervisor agent makes decisions on which agent should be called next.
+- **Supervisor (tool-calling)**: this is a special case of supervisor architecture. Individual agents can be represented as tools. In this case, a supervisor agent uses a tool-calling LLM to decide which of the agent tools to call, as well as the arguments to pass to those agents.
+- **Hierarchical**: you can define a multi-agent system with [a supervisor of supervisors](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/hierarchical_agent_teams/). This is a generalization of the supervisor architecture and allows for more complex control flows.
+- **Custom multi-agent workflow**: each agent communicates with only a subset of agents. Parts of the flow are deterministic, and only some agents can decide which other agents to call next.
 
-Agents can also be defined as tools. In this case, the orchestrator agent (e.g. ReAct agent) would use a tool-calling LLM to decide which of the agent tools to call, as well as the arguments to pass to those agents.
+### Network
 
-You could also take a "mega-graph" approach – incorporating subordinate agents' nodes directly into the parent, orchestrator graph. However, this is not recommended for complex subordinate agents, as it would make the overall system harder to scale, maintain and debug – you should use subgraphs or tools in those cases.
+In this architecture, agents are defined as graph nodes. Each agent can communicate with every other agent (many-to-many connections) and can decide which agent to call next. While very flexible, this architecture doesn't scale well as the number of agents grows:
 
-## Communication in multi-agent systems
+- hard to enforce which agent should be called next
+- hard to determine how much [information](#shared-message-list) should be passed between the agents
 
-A big question in multi-agent systems is how the agents communicate amongst themselves and with the orchestrator agent. This involves both the schema of how they communicate, as well as the sequence in which they communicate. LangGraph is perfect for orchestrating these types of systems and allows you to define both.
+We recommend avoiding this architecture in production and using one of the below architectures instead.
 
-### Schema
+### Supervisor
 
-LangGraph provides a lot of flexibility for how to communicate within multi-agent architectures.
-
-* A node in LangGraph can have a [private input state schema](https://langchain-ai.github.io/langgraph/how-tos/pass_private_state/) that is distinct from the graph state schema. This allows passing additional information during the graph execution that is only needed for executing a particular node.
-* Subgraph node agents can have independent [input / output state schemas](https://langchain-ai.github.io/langgraph/how-tos/input_output_schema/). In this case it’s important to [add input / output transformations](https://langchain-ai.github.io/langgraph/how-tos/subgraph-transform-state/) so that the parent graph knows how to communicate with the subgraphs.
-* For tool-based subordinate agents, the orchestrator determines the inputs based on the tool schema. Additionally, LangGraph allows passing state to individual tools at runtime, so subordinate agents can access parent state, if needed.
-
-### Sequence
-
-LangGraph provides multiple methods to control agent communication sequence:
-
-* **Explicit control flow (graph edges)**: LangGraph allows you to define the control flow of your application (i.e. the sequence of how agents communicate) explicitly, via [graph edges](./low_level.md#edges).
+In this architecture, we define agents as nodes and add a supervisor node (LLM) that decides which agent nodes should be called next. We use [conditional edges](./low_level.md#conditional-edges) to route execution to the appropriate agent node based on supervisor's decision. This architecture also lends itself well to running multiple agents in parallel or using [map-reduce](../how-tos/map-reduce.ipynb) pattern.
 
 ```python
+from typing import Literal
 from langchain_openai import ChatOpenAI
-from langchain_core.messages import SystemMessage
-from langgraph.graph import StateGraph, MessagesState, START, END
+from langgraph.graph import StateGraph, MessagesState, START
 
-model = ChatOpenAI(model="gpt-4o-mini")
+model = ChatOpenAI()
 
-def research_agent(state: MessagesState):
-    """Call research agent"""
-    messages = [SystemMessage(content="You are a research assistant. Given a topic, provide key facts and information.")] + state["messages"]
-    response = model.invoke(messages)
+class AgentState(MessagesState):
+    next: Literal["agent_1", "agent_2", "__end__"]
+
+def supervisor(state: AgentState):
+    # you can pass relevant parts of the state to the LLM (e.g., state["messages"])
+    # to determine which agent to call next. a common pattern is to call the model
+    # with a structured output (e.g. force it to return an output with a "next_agent" field)
+    response = model.invoke(...)
+    # the "next" key will be used by the conditional edges to route execution
+    # to the appropriate agent
+    return {"next": response["next_agent"]}
+
+def agent_1(state: AgentState):
+    # you can pass relevant parts of the state to the LLM (e.g., state["messages"])
+    # and add any additional logic (different models, custom prompts, structured output, etc.)
+    response = model.invoke(...)
     return {"messages": [response]}
 
-def summarize_agent(state: MessagesState):
-    """Call summarization agent"""
-    messages = [SystemMessage(content="You are a summarization expert. Condense the given information into a brief summary.")] + state["messages"]
-    response = model.invoke(messages)
+def agent_2(state: AgentState):
+    response = model.invoke(...)
     return {"messages": [response]}
 
-graph = StateGraph(MessagesState)
-graph.add_node("research", research_agent)
-graph.add_node("summarize", summarize_agent)
+builder = StateGraph(AgentState)
+builder.add_node(supervisor)
+builder.add_node(agent_1)
+builder.add_node(agent_2)
 
-# define the flow explicitly
-graph.add_edge(START, "research")
-graph.add_edge("research", "summarize")
-graph.add_edge("summarize", END)
+builder.add_edge(START, "supervisor")
+# route to one of the agents or exit based on the supervisor's decisiion
+# if the supervisor returns "__end__", the graph will finish execution
+builder.add_conditional_edges("supervisor", lambda state: state["next"])
+builder.add_edge("agent_1", "supervisor")
+builder.add_edge("agent_2", "supervisor")
+
+supervisor = builder.compile()
 ```
 
-* **Dynamic control flow (conditional edges)**: LangGraph also allows you to define [conditional edges](./low_level.md#conditional-edges), where the control flow is dependent on satisfying a given condition. In such cases, you can use an LLM to decide which subordinate agent to call next.
+Check out this [tutorial](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/agent_supervisor/) for an example of supervisor multi-agent architecture.
 
+### Supervisor (tool-calling)
 
-* **Implicit control flow (tool calling)**: if the orchestrator agent treats subordinate agents as tools, the tool-calling LLM powering the orchestrator will make decisions about the order in which the tools (agents) are being called.
+In this variant of the [supervisor](#supervisor) architecture, we define individual agents as **tools** and use a tool-calling LLM in the supervisor node. This can be implemented as a [ReAct](./agentic_concepts.md#react-implementation)-style agent with two nodes — an LLM node (supervisor) and a tool-calling node that executes tools (agents in this case).
 
 ```python
 from typing import Annotated
-from langchain_core.messages import SystemMessage, ToolMessage
 from langchain_openai import ChatOpenAI
-from langgraph.prebuilt import ToolNode, InjectedState, create_react_agent
+from langgraph.prebuilt import InjectedState, create_react_agent
 
-model = ChatOpenAI(model="gpt-4o-mini")
+model = ChatOpenAI()
 
-def research_agent(state: Annotated[dict, InjectedState]):
-    """Call research agent"""
-    messages = [SystemMessage(content="You are a research assistant. Given a topic, provide key facts and information.")] + state["messages"][:-1]
-    response = model.invoke(messages)
-    tool_call = state["messages"][-1].tool_calls[0]
-    return {"messages": [ToolMessage(response.content, tool_call_id=tool_call["id"])]}
+# this is the agent function that will be called as tool
+# notice that you can pass the state to the tool via InjectedState annotation
+def agent_1(state: Annotated[dict, InjectedState]):
+    # you can pass relevant parts of the state to the LLM (e.g., state["messages"])
+    # and add any additional logic (different models, custom prompts, structured output, etc.)
+    response = model.invoke(...)
+    # return the LLM response as a string (expected tool response format)
+    # this will be automatically turned to ToolMessage
+    # by the prebuilt create_react_agent (supervisor)
+    return response.content
 
-def summarize_agent(state: Annotated[dict, InjectedState]):
-    """Call summarization agent"""
-    messages = [SystemMessage(content="You are a summarization expert. Condense the given information into a brief summary.")] + state["messages"][:-1]
-    response = model.invoke(messages)
-    tool_call = state["messages"][-1].tool_calls[0]
-    return {"messages": [ToolMessage(response.content, tool_call_id=tool_call["id"])]}
+def agent_2(state: Annotated[dict, InjectedState]):
+    response = model.invoke(...)
+    return response.content
 
-tool_node = ToolNode([research_agent, summarize_agent])
-graph = create_react_agent(model, [research_agent, summarize_agent], state_modifier="First research and then summarize information on a given topic.")
+tools = [agent_1, agent_2]
+# the simplest way to build a supervisor w/ tool-calling is to use prebuilt ReAct agent graph
+# that consists of a tool-calling LLM node (i.e. supervisor) and a tool-executing node
+supervisor = create_react_agent(model, tools)
 ```
 
-## Example architectures
+### Hierarchical
 
-Below are several examples of complex multi-agent architectures that can be implemented in LangGraph.
+As you add more agents to your system, it might become too hard for the supervisor to manage all of them. The supervisor might start making poor decisions about which agent to call next, the context might become too complex for a single supervisor to keep track of. In other words, you end up with the same problems that motivated the multi-agent architecture in the first place.
 
-### Multi-Agent Collaboration
+To address this, you can design your system _hierarchically_. For example, you can create separate, specialized teams of agents managed by individual supervisors, and a top-level supervisor to manage the teams.
 
-In this example, different agents collaborate on a **shared** scratchpad of messages (i.e. shared graph state). This means that all the work any of them do is visible to the other ones. The benefit is that the other agents can see all the individual steps done. The downside is that sometimes is it overly verbose and unnecessary to pass ALL this information along, and sometimes only the final answer from an agent is needed. We call this **collaboration** because of the shared nature the scratchpad.
+```python
+from typing import Literal
+from langchain_openai import ChatOpenAI
+from langgraph.graph import StateGraph, MessagesState, START
 
-In this case, the independent agents are actually just a single LLM call with a custom system message.
+model = ChatOpenAI()
 
-Here is a visualization of how these agents are connected:
+# define team 1 (same as the single supervisor example above)
+class Team1State(MessagesState):
+    next: Literal["team_1_agent_1", "team_1_agent_2", "__end__"]
 
-![](./img/multi_agent/collaboration.png)
+def team_1_supervisor(state: Team1State):
+    response = model.invoke(...)
+    return {"next": response["next_agent"]}
 
-See full code example in this [tutorial](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/multi-agent-collaboration/).
+def team_1_agent_1(state: Team1State):
+    response = model.invoke(...)
+    return {"messages": [response]}
 
-### Agent Supervisor
+def team_1_agent_2(state: Team1State):
+    response = model.invoke(...)
+    return {"messages": [response]}
 
-In this example, multiple agents are connected, but compared to above they do NOT share a shared scratchpad. Rather, they have their own independent scratchpads (i.e. their own state), and then their final responses are appended to a global scratchpad.
+team_1_builder = StateGraph(Team1State)
+team_1_builder.add_node(team_1_supervisor)
+team_1_builder.add_node(team_1_agent_1)
+team_1_builder.add_node(team_1_agent_2)
+team_1_builder.add_edge(START, "team_1_supervisor")
+# route to one of the agents or exit based on the supervisor's decisiion
+# if the supervisor returns "__end__", the graph will finish execution
+team_1_builder.add_conditional_edges("team_1_supervisor", lambda state: state["next"])
+team_1_builder.add_edge("team_1_agent_1", "team_1_supervisor")
+team_1_builder.add_edge("team_1_agent_2", "team_1_supervisor")
 
-In this case, the independent agents are a LangGraph ReAct agent (graph). This means they have their own individual prompt, LLM, and tools. When called, it's not just a single LLM call, but rather an invocation of the graph powering the ReAct agent.
+team_1_graph = team_1_builder.compile()
 
-![](./img/multi_agent/supervisor.png)
+# define team 2 (same as the single supervisor example above)
+class Team2State(MessagesState):
+    next: Literal["team_2_agent_1", "team_2_agent_2", "__end__"]
 
-See full code example in this [tutorial](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/agent_supervisor/).
+def team_2_supervisor(state: Team2State):
+    ...
 
-### Hierarchical Agent Teams
+def team_2_agent_1(state: Team2State):
+    ...
 
-What if the job for a single worker in agent supervisor example becomes too complex? What if the number of workers becomes too large? For some applications, the system may be more effective if work is distributed hierarchically. You can do this by creating additional level of subgraphs and creating a top-level supervisor, along with mid-level supervisors:
+def team_2_agent_2(state: Team2State):
+    ...
 
-![](./img/multi_agent/hierarchical.png)
+team_2_builder = StateGraph(Team2State)
+...
+team_2_graph = team_2_builder.compile()
 
-See full code example in this [tutorial](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/hierarchical_agent_teams/).
+
+# define top-level supervisor
+
+class TopLevelState(MessagesState):
+    next: Literal["team_1", "team_2", "__end__"]
+
+builder = StateGraph(TopLevelState)
+def top_level_supervisor(state: TopLevelState):
+    # you can pass relevant parts of the state to the LLM (e.g., state["messages"])
+    # to determine which team to call next. a common pattern is to call the model
+    # with a structured output (e.g. force it to return an output with a "next_team" field)
+    response = model.invoke(...)
+    # the "next" key will be used by the conditional edges to route execution
+    # to the appropriate team
+    return {"next": response["next_team"]}
+
+builder = StateGraph(TopLevelState)
+builder.add_node(top_level_supervisor)
+builder.add_node(team_1_graph)
+builder.add_node(team_2_graph)
+
+builder.add_edge(START, "top_level_supervisor")
+# route to one of the teams or exit based on the supervisor's decision
+# if the top-level supervisor returns "__end__", the graph will finish execution
+builder.add_conditional_edges("top_level_supervisor", lambda state: state["next"])
+builder.add_edge("team_1_graph", "top_level_supervisor")
+builder.add_edge("team_2_graph", "top_level_supervisor")
+
+graph = builder.compile()
+```
+
+### Custom multi-agent workflow
+
+In this architecture we add individual agents as graph nodes and define the order in which agents are called ahead of time, in a custom workflow. In LangGraph the workflow can be defined in two ways:
+
+- **Explicit control flow (normal edges)**: LangGraph allows you to explicitly define the control flow of your application (i.e. the sequence of how agents communicate) explicitly, via [normal graph edges](./low_level.md#normal-edges). This is the most deterministic variant of this architecture above — we always know which agent will be called next ahead of time.
+
+- **Dynamic control flow (conditional edges)**: in LangGraph you can allow LLMs to decide parts of your application control flow. This can be achieved by using [conditional edges](./low_level.md#conditional-edges). A special case of this is a [supervisor tool-calling](#supervisor-tool-calling) architecture. In that case, the tool-calling LLM powering the supervisor agent will make decisions about the order in which the tools (agents) are being called.
+
+```python
+from langchain_openai import ChatOpenAI
+from langgraph.graph import StateGraph, MessagesState, START
+
+model = ChatOpenAI()
+
+def agent_1(state: MessagesState):
+    response = model.invoke(...)
+    return {"messages": [response]}
+
+def agent_2(state: MessagesState):
+    response = model.invoke(...)
+    return {"messages": [response]}
+
+builder = StateGraph(MessagesState)
+builder.add_node(agent_1)
+builder.add_node(agent_2)
+# define the flow explicitly
+builder.add_edge(START, "agent_1")
+builder.add_edge("agent_1", "agent_2")
+```
+
+## Communication between agents
+
+The most important thing when building multi-agent systems is figuring out how the agents communicate. There are few different considerations:
+
+- Do agents communicate via [**via graph state or via tool calls**](#graph-state-vs-tool-calls)?
+- What if two agents have [**different state schemas**](#different-state-schemas)?
+- How to communicate over a [**shared message list**](#shared-message-list)?
+
+### Graph state vs tool calls
+
+What is the "payload" that is being passed around between agents? In most of the architectures discussed above the agents communicate via the [graph state](./low_level.md#state). In the case of the [supervisor with tool-calling](#supervisor-tool-calling), the payloads are tool call arguments.
+
+![](./img/multi_agent/request.png)
+
+#### Graph state
+
+To communicate via graph state, individual agents need to be defined as [graph nodes](./low_level.md#nodes). These can be added as functions or as entire [subgraphs](./low_level.md#subgraphs). At each step of the graph execution, agent node receives the current state of the graph, executes the agent code and then passes the updated state to the next nodes.
+
+Typically agent nodes share a single [state schema](./low_level.md#schema). However, you might want to design agent nodes with [different state schemas](#different-state-schemas).
+
+### Different state schemas
+
+An agent might need to have a different state schema from the rest of the agents. For example, a search agent might only need to keep track of queries and retrieved documents. There are two ways to achieve this in LangGraph:
+
+- Define [subgraph](./low_level.md#subgraphs) agents with a separate state schema. If there are no shared state keys (channels) between the subgraph and the parent graph, it’s important to [add input / output transformations](https://langchain-ai.github.io/langgraph/how-tos/subgraph-transform-state/) so that the parent graph knows how to communicate with the subgraphs.
+- Define agent node functions with a [private input state schema](https://langchain-ai.github.io/langgraph/how-tos/pass_private_state/) that is distinct from the overall graph state schema. This allows passing information that is only needed for executing that particular agent.
+
+### Shared message list
+
+The most common way for the agents to communicate is via a shared state channel, typically a list of messages. This assumes that there is always at least a single channel (key) in the state that is shared by the agents. When communicating via a shared message list there is an additional consideration: should the agents [share the full history](#share-full-history) of their thought process or only [the final result](#share-final-result)?
+
+![](./img/multi_agent/response.png)
+
+#### Share full history
+
+Agents can **share the full history** of their thought process (i.e. "scratchpad") with all other agents. This "scratchpad" would typically look like a [list of messages](./low_level.md#why-use-messages). The benefit of sharing full thought process is that it might help other agents make better decisions and improve reasoning ability for the system as a whole. The downside is that as the number of agents and their complexity grows, the "scratchpad" will grow quickly and might require additional strategies for [memory management](./memory.md/#managing-long-conversation-history).
+
+#### Share final result
+
+Agents can have their own private "scratchpad" and only **share the final result** with the rest of the agents. This approach might work better for systems with many agents or agents that are more complex. In this case, you would need to define agents with [different state schemas](#different-state-schemas)
+
+For agents called as tools, the supervisor determines the inputs based on the tool schema. Additionally, LangGraph allows [passing state](https://langchain-ai.github.io/langgraph/how-tos/pass-run-time-values-to-tools/#pass-graph-state-to-tools) to individual tools at runtime, so subordinate agents can access parent state, if needed.

docs/docs/concepts/persistence.md

@@ -216,6 +216,148 @@ The final thing you can optionally specify when calling `update_state` is `as_no
 
 ![Update](img/persistence/checkpoints_full_story.jpg)
 
+## Memory Store
+
+![Update](img/persistence/shared_state.png)
+
+A [state schema](low_level.md#schema) specifies a set of keys that are populated as a graph is executed. As discussed above, state can be written by a checkpointer to a thread at each graph step, enabling state persistence.
+
+But, what if we want to retrain some information *across threads*? Consider the case of a chatbot where we want to retain specific information about the user across *all* chat conversations (e.g., threads) with that user!
+
+With checkpointers alone, we cannot share information across threads. This motivates the need for the `Store` interface. As an illustration, we can define an `InMemoryStore` to store information about a user across threads. We simply compile our graph with a checkpointer, as before, and will our new `in_memory_store`.
+First, let's showcase this in isolation without using LangGraph.
+
+```python
+from langgraph.store.memory import InMemoryStore
+in_memory_store = InMemoryStore()
+```
+
+Memories are namespaced by a `tuple`, which in this specific example will be `(<user_id>, "memories")`. The namespace can be any length and represent anything, does not have be user specific.
+
+```python 
+user_id = "1"
+namespace_for_memory = (user_id, "memories")
+```
+
+We use the `store.put` to save memories to our namespace in the store. When we do this, we specify the namespace, as defined above, and a key-value pair for the memory: the key is simply a unique identifier for the memory (`memory_id`) and the value (a dictionary) is the memory itself.
+
+```python
+memory_id = str(uuid.uuid4())
+memory = {"food_preference" : "I like pizza"}
+in_memory_store.put(namespace_for_memory, memory_id, memory)
+```
+
+We can read out memories in our namespace using `store.search`, which will return all memories for a given user as a list. The most recent memory is the last in the list.
+
+```python
+memories = in_memory_store.search(namespace_for_memory)
+memories[-1].dict()
+{'value': {'food_preference': 'I like pizza'},
+ 'key': '07e0caf4-1631-47b7-b15f-65515d4c1843',
+ 'namespace': ['1', 'memories'],
+ 'created_at': '2024-10-02T17:22:31.590602+00:00',
+ 'updated_at': '2024-10-02T17:22:31.590605+00:00'}
+```
+
+Each memory type is a Python class with certain attributes. We can access it as a dictionary by converting via `.dict` as above.
+The attributes it has are:
+
+- `value`: The value (itself a dictionary) of this memory
+- `key`: The UUID for this memory in this namespace
+- `namespace`: A list of strings, the namespace of this memory type
+- `created_at`: Timestamp for when this memory was created
+- `updated_at`: Timestamp for when this memory was updated
+
+With this all in place, we use the `in_memory_store` in LangGraph. The `in_memory_store` works hand-in-hand with the checkpointer: the checkpointer saves state to threads, as discussed above, and the the `in_memory_store` allows us to store arbitrary information for access *across* threads. We compile the graph with both the checkpointer and the `in_memory_store` as follows. 
+
+```python
+from langgraph.checkpoint.memory import MemorySaver
+
+# We need this because we want to enable threads (conversations)
+checkpointer = MemorySaver()
+
+# ... Define the graph ...
+
+# Compile the graph with the checkpointer and store
+graph = graph.compile(checkpointer=checkpointer, store=in_memory_store)
+```
+
+We invoke the graph with a `thread_id`, as before, and also with a `user_id`, which we'll use to namespace our memories to this particular user as we showed above.
+
+```python
+# Invoke the graph
+user_id = "1"
+config = {"configurable": {"thread_id": "1", "user_id": user_id}}
+
+# First let's just say hi to the AI
+for update in graph.stream(
+    {"messages": [{"role": "user", "content": "hi"}]}, config, stream_mode="updates"
+):
+    print(update)
+```
+
+We can access the `in_memory_store` and the `user_id` in *any node* by passing `store: BaseStore` and `config: RunnableConfig` as node arguments. Just as we saw above, simply use the `put` method to save memories to the store.
+
+```python
+def update_memory(state: MessagesState, config: RunnableConfig, *, store: BaseStore):
+    
+    # Get the user id from the config
+    user_id = config["configurable"]["user_id"]
+    
+    # Namespace the memory
+    namespace = (user_id, "memories")
+    
+    # ... Analyze conversation and create a new memory
+    
+    # Create a new memory ID
+    memory_id = str(uuid.uuid4())
+
+    # We create a new memory
+    store.put(namespace, memory_id, {"memory": memory})
+
+```
+
+As we showed above, we can also access the store in any node and use `search` to get memories. Recall the the memories are returned as a list of objects that can be converted to a dictionary.
+
+```python
+memories[-1].dict()
+{'value': {'food_preference': 'I like pizza'},
+ 'key': '07e0caf4-1631-47b7-b15f-65515d4c1843',
+ 'namespace': ['1', 'memories'],
+ 'created_at': '2024-10-02T17:22:31.590602+00:00',
+ 'updated_at': '2024-10-02T17:22:31.590605+00:00'}
+```
+
+We can access the memories and use them in our model call.
+
+```python
+def call_model(state: MessagesState, config: RunnableConfig, *, store: BaseStore):
+    
+    # Get the user id from the config
+    user_id = config["configurable"]["user_id"]
+    
+    # Get the memories for the user from the store
+    memories = store.search(("memories", user_id))
+    info = "\n".join([d.value["memory"] for d in memories])
+    
+    # ... Use memories in the model call
+```
+
+If we create a new thread, we can still access the same memories so long as the `user_id` is the same. 
+
+```python
+# Invoke the graph
+config = {"configurable": {"thread_id": "2", "user_id": "1"}}
+
+# Let's say hi again
+for update in graph.stream(
+    {"messages": [{"role": "user", "content": "hi, tell me about my memories"}]}, config, stream_mode="updates"
+):
+    print(update)
+```
+
+When we use the LangGraph API, either locally (e.g., in LangGraph Studio) or with LangGraph Cloud, the memory store is available to use by default and does not need to be specified during graph compilation.
+
 ## Checkpointer libraries
 
 Under the hood, checkpointing is powered by checkpointer objects that conform to [BaseCheckpointSaver][langgraph.checkpoint.base.BaseCheckpointSaver] interface. LangGraph provides several checkpointer implementations, all implemented via standalone, installable libraries:

docs/docs/how-tos/async.ipynb

@@ -46,7 +46,7 @@
     "    <p>\n",
     "        In this how-to, we will create our agent from scratch to be transparent (but verbose). You can accomplish similar functionality using the <code>create_react_agent(model, tools=tool)</code> (<a href=\"https://langchain-ai.github.io/langgraph/reference/prebuilt/#create_react_agent\">API doc</a>) constructor. This may be more appropriate if you are used to LangChain’s <a href=\"https://python.langchain.com/v0.1/docs/modules/agents/concepts/#agentexecutor\">AgentExecutor</a> class.\n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {
@@ -107,7 +107,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/branching.ipynb

@@ -72,7 +72,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {
@@ -89,7 +89,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 2,
    "id": "09372b8b-edea-4b9d-9ec3-3d93ce1ba819",
    "metadata": {},
    "outputs": [],
@@ -132,7 +132,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 3,
    "id": "66f52a20",
    "metadata": {},
    "outputs": [
@@ -163,7 +163,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 4,
    "id": "38846b01",
    "metadata": {},
    "outputs": [
@@ -183,7 +183,7 @@
        "{'aggregate': [\"I'm A\", \"I'm B\", \"I'm C\", \"I'm D\"]}"
       ]
      },
-     "execution_count": 5,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -220,7 +220,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 5,
    "id": "259a7704-5aa0-4e4c-aeef-cca04e8be0ff",
    "metadata": {},
    "outputs": [],
@@ -238,6 +238,15 @@
     "    aggregate: Annotated[list, operator.add]\n",
     "\n",
     "\n",
+    "class ReturnNodeValue:\n",
+    "    def __init__(self, node_secret: str):\n",
+    "        self._value = node_secret\n",
+    "\n",
+    "    def __call__(self, state: State) -> Any:\n",
+    "        print(f\"Adding {self._value} to {state['aggregate']}\")\n",
+    "        return {\"aggregate\": [self._value]}\n",
+    "\n",
+    "\n",
     "builder = StateGraph(State)\n",
     "builder.add_node(\"a\", ReturnNodeValue(\"I'm A\"))\n",
     "builder.add_edge(START, \"a\")\n",
@@ -255,7 +264,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 6,
    "id": "83320227-8ab3-44c0-b6cf-064a7a425b9f",
    "metadata": {},
    "outputs": [
@@ -278,7 +287,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 7,
    "id": "3f971fa3-29e4-466f-a85e-2863bfecf7fe",
    "metadata": {},
    "outputs": [
@@ -299,7 +308,7 @@
        "{'aggregate': [\"I'm A\", \"I'm B\", \"I'm C\", \"I'm B2\", \"I'm D\"]}"
       ]
      },
-     "execution_count": 8,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -322,7 +331,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 8,
    "id": "95f5e026",
    "metadata": {},
    "outputs": [],
@@ -341,6 +350,15 @@
     "    which: str\n",
     "\n",
     "\n",
+    "class ReturnNodeValue:\n",
+    "    def __init__(self, node_secret: str):\n",
+    "        self._value = node_secret\n",
+    "\n",
+    "    def __call__(self, state: State) -> Any:\n",
+    "        print(f\"Adding {self._value} to {state['aggregate']}\")\n",
+    "        return {\"aggregate\": [self._value]}\n",
+    "\n",
+    "\n",
     "builder = StateGraph(State)\n",
     "builder.add_node(\"a\", ReturnNodeValue(\"I'm A\"))\n",
     "builder.add_edge(START, \"a\")\n",
@@ -372,7 +390,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 9,
    "id": "1d0e6c56",
    "metadata": {},
    "outputs": [
@@ -395,7 +413,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 10,
    "id": "7134f652",
    "metadata": {},
    "outputs": [
@@ -415,7 +433,7 @@
        "{'aggregate': [\"I'm A\", \"I'm B\", \"I'm C\", \"I'm E\"], 'which': 'bc'}"
       ]
      },
-     "execution_count": 11,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -426,7 +444,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 11,
    "id": "b130e694",
    "metadata": {},
    "outputs": [
@@ -446,7 +464,7 @@
        "{'aggregate': [\"I'm A\", \"I'm C\", \"I'm D\", \"I'm E\"], 'which': 'cd'}"
       ]
      },
-     "execution_count": 12,
+     "execution_count": 11,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -471,7 +489,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 12,
    "id": "836bc12d",
    "metadata": {},
    "outputs": [],
@@ -564,7 +582,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 13,
    "id": "932c497e",
    "metadata": {},
    "outputs": [
@@ -587,7 +605,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 14,
    "id": "933b3afd",
    "metadata": {},
    "outputs": [
@@ -608,7 +626,7 @@
        " 'which': 'bc'}"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 14,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -619,7 +637,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 15,
    "id": "e30531bf",
    "metadata": {},
    "outputs": [
@@ -640,7 +658,7 @@
        " 'which': 'cd'}"
       ]
      },
-     "execution_count": 16,
+     "execution_count": 15,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -666,7 +684,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.12.3"
   }
  },
  "nbformat": 4,

docs/docs/how-tos/configuration.ipynb

@@ -76,7 +76,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/create-react-agent-hitl.ipynb

@@ -91,7 +91,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/create-react-agent-memory.ipynb

@@ -96,7 +96,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/create-react-agent-system-prompt.ipynb

@@ -92,7 +92,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/create-react-agent.ipynb

@@ -100,7 +100,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/cross-thread-persistence.ipynb

@@ -66,7 +66,7 @@
    "metadata": {},
    "outputs": [
     {
-     "name": "stdin",
+     "name": "stdout",
      "output_type": "stream",
      "text": [
       "ANTHROPIC_API_KEY:  ········\n"
@@ -91,12 +91,9 @@
    "id": "51b6817d",
    "metadata": {},
    "source": [
-    "<div class=\"admonition tip\">\n",
-    "    <p class=\"admonition-title\">Set up <a href=\"https://smith.langchain.com\">LangSmith</a> for LangGraph development</p>\n",
-    "    <p style=\"padding-top: 5px;\">\n",
-    "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
-    "    </p>\n",
-    "</div>    "
+    "!!! tip \"Set up [LangSmith](https://smith.langchain.com) for LangGraph development\"\n",
+    "\n",
+    "    Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started [here](https://docs.smith.langchain.com)"
    ]
   },
   {

docs/docs/how-tos/human_in_the_loop/breakpoints.ipynb

@@ -85,7 +85,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/human_in_the_loop/dynamic_breakpoints.ipynb

@@ -45,7 +45,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/human_in_the_loop/edit-graph-state.ipynb

@@ -85,7 +85,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/human_in_the_loop/review-tool-calls.ipynb

@@ -82,7 +82,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/human_in_the_loop/time-travel.ipynb

@@ -89,7 +89,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/human_in_the_loop/wait-user-input.ipynb

@@ -78,7 +78,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/index.md

@@ -1,15 +1,18 @@
 ---
 hide:
-  - toc
+  - navigation
+title: How-to Guides
+description: How to accomplish common tasks in LangGraph
 ---
 
-# How-to guides
+# How-to Guides
 
 Welcome to the LangGraph how-to guides! These guides provide practical, step-by-step instructions for accomplishing key tasks in LangGraph.
 
 ## Controllability
 
-LangGraph is known for being a highly controllable agent framework.
+LangGraph offers a high level of control over the execution of your graph.
+
 These how-to guides show how to achieve that controllability.
 
 - [How to create branches for parallel execution](branching.ipynb)
@@ -18,9 +21,10 @@ These how-to guides show how to achieve that controllability.
 
 ## Persistence
 
-LangGraph makes it easy to persist state across graph runs (thread-level persistence) and across threads (cross-thread persistence). These how-to guides show how to add persistence to your graph.
+[LangGraph Persistence](../concepts/persistence.md) makes it easy to persist state across graph runs (thread-level persistence) and across threads (cross-thread persistence). These how-to guides show how to add persistence to your graph.
 
 - [How to add thread-level persistence to your graph](persistence.ipynb)
+- [How to add thread-level persistence to subgraphs](subgraph-persistence.ipynb)
 - [How to add cross-thread persistence to your graph](cross-thread-persistence.ipynb)
 - [How to use Postgres checkpointer for persistence](persistence_postgres.ipynb)
 - [How to create a custom checkpointer using MongoDB](persistence_mongodb.ipynb)
@@ -36,8 +40,8 @@ LangGraph makes it easy to manage conversation [memory](../concepts/memory.md) i
 
 ## Human in the Loop
 
-One of LangGraph's main benefits is that it makes human-in-the-loop workflows easy.
-These guides cover common examples of that.
+[Human-in-the-loop](../concepts/human_in_the_loop.md) functionality allows
+you to involve humans in the decision-making process of your graph. These how-to guides show how to implement human-in-the-loop workflows in your graph.
 
 - [How to add breakpoints](human_in_the_loop/breakpoints.ipynb)
 - [How to add dynamic breakpoints](human_in_the_loop/dynamic_breakpoints.ipynb)
@@ -48,8 +52,7 @@ These guides cover common examples of that.
 
 ## Streaming
 
-LangGraph is built to be streaming first.
-These guides show how to use different streaming modes.
+[Streaming](../concepts/streaming.md) is crucial for enhancing the responsiveness of applications built on LLMs. By displaying output progressively, even before a complete response is ready, streaming significantly improves user experience (UX), particularly when dealing with the latency of LLMs.
 
 - [How to stream full state of your graph](stream-values.ipynb)
 - [How to stream state updates of your graph](stream-updates.ipynb)
@@ -65,6 +68,10 @@ These guides show how to use different streaming modes.
 
 ## Tool calling
 
+[Tool calling](https://python.langchain.com/docs/concepts/tool_calling/) is a type of chat model API that accepts tool schemas, along with messages, as input and returns invocations of those tools as part of the output message. 
+
+These how-to guides show common patterns for tool calling with LangGraph:
+
 - [How to call tools using ToolNode](tool-calling.ipynb)
 - [How to handle tool calling errors](tool-calling-errors.ipynb)
 - [How to pass runtime values to tools](pass-run-time-values-to-tools.ipynb)
@@ -73,8 +80,10 @@ These guides show how to use different streaming modes.
 
 ## Subgraphs
 
-- [How to create subgraphs](subgraph.ipynb)
-- [How to manage state in subgraphs](subgraphs-manage-state.ipynb)
+[Subgraphs](../concepts/low_level.md#subgraphs) allow you to reuse an existing graph from another graph. These how-to guides show how to use subgraphs:
+
+- [How to add and use subgraphs](subgraph.ipynb)
+- [How to view and update state in subgraphs](subgraphs-manage-state.ipynb)
 - [How to transform inputs and outputs of a subgraph](subgraph-transform-state.ipynb)
 
 ## State Management
@@ -96,11 +105,25 @@ These guides show how to use different streaming modes.
 
 ## Prebuilt ReAct Agent
 
-These guides show how to use the prebuilt ReAct agent.
-Please note that here will we use a **prebuilt agent**. One of the big benefits of LangGraph is that you can easily create your own agent architectures. So while it's fine to start here to build an agent quickly, we would strongly recommend learning how to build your own agent so that you can take full advantage of LangGraph.
+The LangGraph [prebuilt ReAct agent](../reference/prebuilt.md#langgraph.prebuilt.chat_agent_executor.create_react_agent) is pre-built implementation of a [tool calling agent](../concepts/agentic_concepts.md#tool-calling-agent).
+
+One of the big benefits of LangGraph is that you can easily create your own agent architectures. So while it's fine to start here to build an agent quickly, we would strongly recommend learning how to build your own agent so that you can take full advantage of LangGraph.
+
+These guides show how to use the prebuilt ReAct agent:
 
 - [How to create a ReAct agent](create-react-agent.ipynb)
 - [How to add memory to a ReAct agent](create-react-agent-memory.ipynb)
 - [How to add a custom system prompt to a ReAct agent](create-react-agent-system-prompt.ipynb)
 - [How to add human-in-the-loop processes to a ReAct agent](create-react-agent-hitl.ipynb)
-- [How to create prebuilt ReAct agent from scratch](react-agent-from-scratch.ipynb)
+- [How to create prebuilt ReAct agent from scratch](react-agent-from-scratch.ipynb)
+
+## Troubleshooting
+
+The [Error Reference](../troubleshooting/errors/index.md) page contains guides around resolving common errors you may find while building with LangChain. Errors referenced below will have an `lc_error_code` property corresponding to one of the below codes when they are thrown in code.
+
+- [GRAPH_RECURSION_LIMIT](../troubleshooting/errors/GRAPH_RECURSION_LIMIT.md)
+- [INVALID_CONCURRENT_GRAPH_UPDATE](../troubleshooting/errors/INVALID_CONCURRENT_GRAPH_UPDATE.md)
+- [INVALID_GRAPH_NODE_RETURN_VALUE](../troubleshooting/errors/INVALID_GRAPH_NODE_RETURN_VALUE.md)
+- [MULTIPLE_SUBGRAPHS](../troubleshooting/errors/MULTIPLE_SUBGRAPHS.md)
+
+

docs/docs/how-tos/input_output_schema.ipynb

@@ -58,7 +58,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/map-reduce.ipynb

@@ -103,7 +103,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/memory/add-summary-conversation-history.ipynb

@@ -75,7 +75,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/memory/delete-messages.ipynb

@@ -82,7 +82,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/memory/manage-conversation-history.ipynb

@@ -73,7 +73,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/node-retries.ipynb

@@ -66,7 +66,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/pass-config-to-tools.ipynb

@@ -100,7 +100,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/pass-run-time-values-to-tools.ipynb

@@ -158,7 +158,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/pass_private_state.ipynb

@@ -51,7 +51,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/persistence.ipynb

@@ -47,7 +47,7 @@
     "<div class=\"admonition tip\">\n",
     "    <p class=\"admonition-title\">Note</p>\n",
     "    <p>\n",
-    "        If you need memory that is <b>shared</b> across multiple conversations or users (cross-thread persistence), check out this <a href=\"https://langchain-ai.github.io/langgraph/how-tos/cross-thread-persistence/\">how-to guide</a>).\n",
+    "        If you need memory that is <b>shared</b> across multiple conversations or users (cross-thread persistence), check out this <a href=\"https://langchain-ai.github.io/langgraph/how-tos/cross-thread-persistence/\">how-to guide</a>.\n",
     "    </p>\n",
     "</div>"
    ]
@@ -88,7 +88,7 @@
    "metadata": {},
    "outputs": [
     {
-     "name": "stdin",
+     "name": "stdout",
      "output_type": "stream",
      "text": [
       "ANTHROPIC_API_KEY:  ········\n"
@@ -118,7 +118,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/persistence_mongodb.ipynb

@@ -102,7 +102,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/persistence_postgres.ipynb

@@ -98,7 +98,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/persistence_redis.ipynb

@@ -102,7 +102,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/react-agent-from-scratch.ipynb

@@ -250,7 +250,6 @@
    ],
    "source": [
     "from langgraph.graph import StateGraph, END\n",
-    "from langgraph.constants import END\n",
     "\n",
     "# Define a new graph\n",
     "workflow = StateGraph(AgentState)\n",

docs/docs/how-tos/react-agent-structured-output.ipynb

@@ -123,7 +123,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/recursion-limit.ipynb

@@ -59,7 +59,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/return-when-recursion-limit-hits.ipynb

@@ -44,7 +44,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -61,7 +61,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {
@@ -70,12 +70,12 @@
    "source": [
     "## Without returning state\n",
     "\n",
-    "We are going to define a dummy graph in this example that will always hit the recursion limit. First, we will implement it without returning the state and show that it hits the recursion limit. This graph is based on the ReACT architecture, but instead of actually making decisions and taking actions it just loops forever."
+    "We are going to define a dummy graph in this example that will always hit the recursion limit. First, we will implement it without returning the state and show that it hits the recursion limit. This graph is based on the ReAct architecture, but instead of actually making decisions and taking actions it just loops forever."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -116,7 +116,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
@@ -145,7 +145,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -171,18 +171,18 @@
    "source": [
     "## With returning state\n",
     "\n",
-    "If we wanted to actually return the state, what we are going to do is introduce a new key to our state called `is_last_step` which keeps track of if we are on the last step of our recursion limit. If so, we will bypass all other graph decisions and simply terminate the graph, returning the state to the user without causing an error.\n",
+    "To avoid hitting the recursion limit, we can introduce a new key to our state called `remaining_steps`. It will keep track of number of steps until reaching the recursion limit. We can then check the value of `remaining_steps` to determine whether we should terminate the graph execution and return the state to the user without causing the `RecursionError`.\n",
     "\n",
-    "We are going to use a `ManagedValue` channel to do this. A `ManagedValue` channel is a state channel that will exist for the duration of our graph run and no longer. Since our `action` node is going to always induce at least 2 extra steps to our graph (since the `action` node ALWAYS calls the `decision` node afterwards), we will use this channel to check if we are within 2 steps of the limit. See the implementation of `IsLastOrSecondToLastStepManager` below.\n",
+    "To do so, we will use a special `RemainingSteps` annotation. Under the hood, it creates a special `ManagedValue` channel -- a state channel that will exist for the duration of our graph run and no longer.\n",
     "\n",
-    "This implementation very closely mirrors the implementation of `isLastStep` (which you can use by calling `from langgraph.managed import IsLastStep` and then decorating state keys with the `isLastStep` type), but in this case we check if we are on the last OR second-to-last step, instead of just the last step.\n",
+    "Since our `action` node is going to always induce at least 2 extra steps to our graph (since the `action` node ALWAYS calls the `decision` node afterwards), we will use this channel to check if we are within 2 steps of the limit.\n",
     "\n",
     "Now, when we run our graph we should receive no errors and instead get the last value of the state before the recursion limit was hit."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -190,24 +190,18 @@
     "from langgraph.graph import StateGraph\n",
     "from typing import Annotated\n",
     "\n",
-    "from langgraph.managed.base import ManagedValue\n",
-    "\n",
-    "\n",
-    "class IsLastOrSecondToLastStepManager(ManagedValue[bool]):\n",
-    "    def __call__(self, step: int) -> bool:\n",
-    "        limit = self.config.get(\"recursion_limit\", 0)\n",
-    "        return step >= limit - 2\n",
+    "from langgraph.managed.is_last_step import RemainingSteps\n",
     "\n",
     "\n",
     "class State(TypedDict):\n",
     "    value: str\n",
     "    action_result: str\n",
-    "    is_last_step: Annotated[bool, IsLastOrSecondToLastStepManager]\n",
+    "    remaining_steps: RemainingSteps\n",
     "\n",
     "\n",
     "def router(state: State):\n",
-    "    # Force the agent to end if it is on the last step\n",
-    "    if state[\"is_last_step\"]:\n",
+    "    # Force the agent to end\n",
+    "    if state[\"remaining_steps\"] <= 2:\n",
     "        return END\n",
     "    if state[\"value\"] == \"end\":\n",
     "        return END\n",
@@ -235,7 +229,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -244,7 +238,7 @@
        "{'value': 'keep going!', 'action_result': 'what a great result!'}"
       ]
      },
-     "execution_count": 5,
+     "execution_count": 6,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -277,7 +271,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.11.9"
   }
  },
  "nbformat": 4,

docs/docs/how-tos/run-id-langsmith.ipynb

@@ -112,7 +112,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/state-model.ipynb

@@ -109,7 +109,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/stream-multiple.ipynb

@@ -66,7 +66,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/stream-updates.ipynb

@@ -71,7 +71,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/stream-values.ipynb

@@ -71,7 +71,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/streaming-content.ipynb

@@ -72,7 +72,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/streaming-events-from-within-tools-without-langchain.ipynb

@@ -89,7 +89,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/streaming-events-from-within-tools.ipynb

@@ -93,7 +93,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/streaming-subgraphs.ipynb

@@ -50,7 +50,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/streaming-tokens-without-langchain.ipynb

@@ -66,7 +66,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/streaming-tokens.ipynb

@@ -16,7 +16,7 @@
     "    <p>\n",
     "        In this how-to, we will create our agent from scratch to be transparent (but verbose). You can accomplish similar functionality using the <code>create_react_agent(model, tools=tool)</code> (<a href=\"https://langchain-ai.github.io/langgraph/reference/prebuilt/#create_react_agent\">API doc</a>) constructor. This may be more appropriate if you are used to LangChain’s <a href=\"https://python.langchain.com/docs/how_to/agent_executor/#concepts\">AgentExecutor</a> class.\n",
     "    </p>\n",
-    "</div>    \n",
+    "</div>\n",
     "\n",
     "<div class=\"admonition warning\">\n",
     "    <p class=\"admonition-title\">Note on Python < 3.11</p>\n",
@@ -24,7 +24,7 @@
     "        When using python 3.8, 3.9, or 3.10, please ensure you manually pass the RunnableConfig through to the llm when invoking it like so: <code>llm.ainvoke(..., config)</code>.\n",
     "        The stream method collects all events from your nested code using a streaming tracer passed as a callback. In 3.11 and above, this is automatically handled via <a href=\"https://docs.python.org/3/library/contextvars.html\">contextvar</a>'s; prior to 3.11, <a href=\"https://docs.python.org/3/library/asyncio-task.html#asyncio.create_task\">asyncio's tasks</a> lacked proper contextvar support, meaning that the callbacks will only propagate if you manually pass the config through. We do this in the <code>call_model</code> method below.\n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {
@@ -85,7 +85,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {
@@ -264,7 +264,7 @@
     "    <p class=\"admonition-title\">Manual Callback Propagation</p>\n",
     "    <p>\n",
     "        Note that in <code>call_model(state: State, config: RunnableConfig):</code> below, we a) accept the <a href=\"https://api.python.langchain.com/en/latest/runnables/langchain_core.runnables.config.RunnableConfig.html#langchain_core.runnables.config.RunnableConfig\">RunnableConfig</a> in the node and b) pass this in as the second arg for <code>llm.ainvoke(..., config)</code>. This is optional for python 3.11 and later.</p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/subgraph-persistence.ipynb

@@ -0,0 +1,379 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "176e8dbb-1a0a-49ce-a10e-2417e8ea17a0",
+   "metadata": {},
+   "source": [
+    "# How to add thread-level persistence to subgraphs"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8c67581a-49fb-4597-a7fc-6774581c2160",
+   "metadata": {},
+   "source": [
+    "<div class=\"admonition tip\">\n",
+    "    <p class=\"admonition-title\">Prerequisites</p>\n",
+    "    <p>\n",
+    "        This guide assumes familiarity with the following:\n",
+    "        <ul>\n",
+    "            <li>            \n",
+    "                <a href=\"https://langchain-ai.github.io/langgraph/concepts/low_level/#subgraphs\">\n",
+    "                    Subgraphs\n",
+    "                </a>\n",
+    "            </li>\n",
+    "            <li>\n",
+    "                <a href=\"https://langchain-ai.github.io/langgraph/concepts/persistence/\">\n",
+    "                    Persistence\n",
+    "                </a>\n",
+    "            </li>\n",
+    "        </ul>\n",
+    "    </p>\n",
+    "</div>\n",
+    "\n",
+    "This guide shows how you can add [thread-level](https://langchain-ai.github.io/langgraph/how-tos/persistence/) persistence to graphs that use [subgraphs](https://langchain-ai.github.io/langgraph/how-tos/subgraph/)."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8f83b855-ab23-4de7-9559-702cad9a29c6",
+   "metadata": {},
+   "source": [
+    "## Setup\n",
+    "\n",
+    "First, let's install the required packages"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "77d1eafa-3252-45f6-9af0-d94e1f9c5c9e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%%capture --no-stderr\n",
+    "%pip install -U langgraph"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "2e60c6cd-bf4e-46af-9761-b872d0fbe3b6",
+   "metadata": {},
+   "source": [
+    "<div class=\"admonition tip\">\n",
+    "    <p class=\"admonition-title\">Set up <a href=\"https://smith.langchain.com\">LangSmith</a> for LangGraph development</p>\n",
+    "    <p style=\"padding-top: 5px;\">\n",
+    "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
+    "    </p>\n",
+    "</div>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "871b9056-fec7-4683-8c22-f56c91f5b13b",
+   "metadata": {},
+   "source": [
+    "## Define the graph with persistence"
+   ]
+  },
+  {
+   "attachments": {},
+   "cell_type": "markdown",
+   "id": "9f1303ef-df37-48e0-8a59-8ff169c52c5b",
+   "metadata": {},
+   "source": [
+    "To add persistence to a graph with subgraphs, all you need to do is pass a [checkpointer](https://langchain-ai.github.io/langgraph/reference/checkpoints/#langgraph.checkpoint.base.BaseCheckpointSaver) when **compiling the parent graph**. LangGraph will automatically propagate the checkpointer to the child subgraphs."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c74cde2e-c127-4326-8d36-b6acef987f0a",
+   "metadata": {},
+   "source": [
+    "!!! note\n",
+    "    You **shouldn't provide** a checkpointer when compiling a subgraph. Instead, you must define a **single** checkpointer that you pass to `parent_graph.compile()`, and LangGraph will automatically propagate the checkpointer to the child subgraphs. If you pass the checkpointer to the `subgraph.compile()`, it will simply be ignored. This also applies when you [add a node function that invokes the subgraph](../subgraph#add-a-node-function-that-invokes-the-subgraph)."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c3a1fe22-1ca9-45eb-a35b-71b9c905e8c5",
+   "metadata": {},
+   "source": [
+    "Let's define a simple graph with a single subgraph node to show how to do this."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "0d76f0c0-bd77-4eca-9527-27bcdf85dd42",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<langgraph.graph.state.StateGraph at 0x106d2fa10>"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from langgraph.graph import START, StateGraph\n",
+    "from langgraph.checkpoint.memory import MemorySaver\n",
+    "from typing import TypedDict\n",
+    "\n",
+    "\n",
+    "# subgraph\n",
+    "\n",
+    "\n",
+    "class SubgraphState(TypedDict):\n",
+    "    foo: str  # note that this key is shared with the parent graph state\n",
+    "    bar: str\n",
+    "\n",
+    "\n",
+    "def subgraph_node_1(state: SubgraphState):\n",
+    "    return {\"bar\": \"bar\"}\n",
+    "\n",
+    "\n",
+    "def subgraph_node_2(state: SubgraphState):\n",
+    "    # note that this node is using a state key ('bar') that is only available in the subgraph\n",
+    "    # and is sending update on the shared state key ('foo')\n",
+    "    return {\"foo\": state[\"foo\"] + state[\"bar\"]}\n",
+    "\n",
+    "\n",
+    "subgraph_builder = StateGraph(SubgraphState)\n",
+    "subgraph_builder.add_node(subgraph_node_1)\n",
+    "subgraph_builder.add_node(subgraph_node_2)\n",
+    "subgraph_builder.add_edge(START, \"subgraph_node_1\")\n",
+    "subgraph_builder.add_edge(\"subgraph_node_1\", \"subgraph_node_2\")\n",
+    "subgraph = subgraph_builder.compile()\n",
+    "\n",
+    "\n",
+    "# parent graph\n",
+    "\n",
+    "\n",
+    "class State(TypedDict):\n",
+    "    foo: str\n",
+    "\n",
+    "\n",
+    "def node_1(state: State):\n",
+    "    return {\"foo\": \"hi! \" + state[\"foo\"]}\n",
+    "\n",
+    "\n",
+    "builder = StateGraph(State)\n",
+    "builder.add_node(\"node_1\", node_1)\n",
+    "# note that we're adding the compiled subgraph as a node to the parent graph\n",
+    "builder.add_node(\"node_2\", subgraph)\n",
+    "builder.add_edge(START, \"node_1\")\n",
+    "builder.add_edge(\"node_1\", \"node_2\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "47084b1f-9fd5-40a9-9d75-89eb5f853d02",
+   "metadata": {},
+   "source": [
+    "We can now compile the graph with an in-memory checkpointer (`MemorySaver`)."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "7657d285-c896-40c9-a569-b4a3b9c230c7",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "checkpointer = MemorySaver()\n",
+    "# You must only pass checkpointer when compiling the parent graph.\n",
+    "# LangGraph will automatically propagate the checkpointer to the child subgraphs.\n",
+    "graph = builder.compile(checkpointer=checkpointer)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "0d193e3c-4ec3-4034-beed-8e5550c6542c",
+   "metadata": {},
+   "source": [
+    "## Verify persistence works"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "eb69a5f0-b92e-4d4e-9aa9-c4c4ec7de91a",
+   "metadata": {},
+   "source": [
+    "Let's now run the graph and inspect the persisted state for both the parent graph and the subgraph to verify that persistence works. We should expect to see the final execution results for both the parent and subgraph in `state.values`."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "13da686e-6ed6-4b83-93e8-1631fcc8c2a9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "config = {\"configurable\": {\"thread_id\": \"1\"}}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "8721f045-2e82-4bf0-9d85-5ba6ecf899d6",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{'node_1': {'foo': 'hi! foo'}}\n",
+      "{'subgraph_node_1': {'bar': 'bar'}}\n",
+      "{'subgraph_node_2': {'foo': 'hi! foobar'}}\n",
+      "{'node_2': {'foo': 'hi! foobar'}}\n"
+     ]
+    }
+   ],
+   "source": [
+    "for _, chunk in graph.stream({\"foo\": \"foo\"}, config, subgraphs=True):\n",
+    "    print(chunk)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "ec6b5ce4-becc-4910-8a6d-d6b60d9d6f60",
+   "metadata": {},
+   "source": [
+    "We can now view the parent graph state by calling `graph.get_state()` with the same config that we used to invoke the graph."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "3e817283-142d-4fda-8cb1-8de34717f833",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'foo': 'hi! foobar'}"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "graph.get_state(config).values"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "fbc4f30b-941e-4140-8bfa-3b8cc670489c",
+   "metadata": {},
+   "source": [
+    "To view the subgraph state, we need to do two things:\n",
+    "\n",
+    "1. Find the most recent config value for the subgraph\n",
+    "2. Use `graph.get_state()` to retrieve that value for the most recent subgraph config.\n",
+    "\n",
+    "To find the correct config, we can examine the state history from the parent graph and find the state snapshot before we return results from `node_2` (the node with subgraph):"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "e896628f-36b2-45eb-b7c5-c64c1098f328",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "state_with_subgraph = [\n",
+    "    s for s in graph.get_state_history(config) if s.next == (\"node_2\",)\n",
+    "][0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7af49977-42b1-40a1-88f1-f07437f8b7f9",
+   "metadata": {},
+   "source": [
+    "The state snapshot will include the list of `tasks` to be executed next. When using subgraphs, the `tasks` will contain the config that we can use to retrieve the subgraph state:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "21e96df3-946d-40f8-8d6d-055ae4177452",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'configurable': {'thread_id': '1',\n",
+       "  'checkpoint_ns': 'node_2:6ef111a6-f290-7376-0dfc-a4152307bc5b'}}"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "subgraph_config = state_with_subgraph.tasks[0].state\n",
+    "subgraph_config"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "1d2401b3-d52b-4895-a5d1-dccf015ba216",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'foo': 'hi! foobar', 'bar': 'bar'}"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "graph.get_state(subgraph_config).values"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "40aded92-99dd-427b-932d-aa78f474c271",
+   "metadata": {},
+   "source": [
+    "If you want to learn more about how to modify the subgraph state for human-in-the-loop workflows, check out this [how-to guide](https://langchain-ai.github.io/langgraph/how-tos/subgraphs-manage-state/)."
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

docs/docs/how-tos/subgraph-transform-state.ipynb

@@ -35,7 +35,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {
@@ -171,7 +171,7 @@
     "    <p>\n",
     "    We're wrapping the <code>grandchild_graph</code> invocation in a separate function (<code>call_grandchild_graph</code>) that transforms the input state before calling the grandchild graph and then transforms the output of grandchild graph back to child graph state. If you just pass <code>grandchild_graph</code> directly to <code>.add_node</code> without the transformations, LangGraph will raise an error as there are no shared state channels (keys) between child and grandchild states.\n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {
@@ -240,7 +240,7 @@
     "    <p>\n",
     "    We're wrapping the <code>child_graph</code> invocation in a separate function (<code>call_child_graph</code>) that transforms the input state before calling the child graph and then transforms the output of the child graph back to parent graph state. If you just pass <code>child_graph</code> directly to <code>.add_node</code> without the transformations, LangGraph will raise an error as there are no shared state channels (keys) between parent and child states.\n",
     "    </p>\n",
-    "</div>    \n",
+    "</div>\n",
     "\n",
     "Let's run the parent graph and make sure it correctly calls both the child and grandchild subgraphs:"
    ]

docs/docs/how-tos/subgraphs-manage-state.ipynb

@@ -5,15 +5,48 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "# How to manage state in subgraphs\n",
+    "# How to view and update state in subgraphs\n",
     "\n",
-    "For more complex systems, sub-graphs are a useful design principle. Sub-graphs allow you to create and manage different states in different parts of your graph. This allows you build things like [multi-agent teams](https://langchain-ai.github.io/langgraph/tutorials/multi_agent/hierarchical_agent_teams/), where each team can track its own separate state.\n",
+    "<div class=\"admonition tip\">\n",
+    "    <p class=\"admonition-title\">Prerequisites</p>\n",
+    "    <p>\n",
+    "        This guide assumes familiarity with the following:\n",
+    "        <ul>\n",
+    "            <li>            \n",
+    "                <a href=\"https://langchain-ai.github.io/langgraph/concepts/low_level/#subgraphs\">\n",
+    "                    Subgraphs\n",
+    "                </a>\n",
+    "            </li>\n",
+    "            <li>            \n",
+    "                <a href=\"https://langchain-ai.github.io/langgraph/concepts/human_in_the_loop/\">\n",
+    "                    Human-in-the-loop\n",
+    "                </a>\n",
+    "            </li>\n",
+    "            <li>            \n",
+    "                <a href=\"https://langchain-ai.github.io/langgraph/concepts/low_level/#state\">\n",
+    "                    State\n",
+    "                </a>\n",
+    "            </li>\n",
+    "        </ul>\n",
+    "    </p>\n",
+    "</div> \n",
     "\n",
-    "In this how-to guide we will cover how to manage the persisted state in subgraphs. This will enable a lot of the human-in-the-loop interaction patterns.\n",
+    "Once you add [persistence](../subgraph-persistence), you can easily view and update the state of the subgraph at any point in time. This enables a lot of the human-in-the-loop interaction patterns:\n",
     "\n",
+    "* You can surface a state during an interrupt to a user to let them accept an action.\n",
+    "* You can rewind the subgraph to reproduce or avoid issues.\n",
+    "* You can modify the state to let the user better control its actions.\n",
+    "\n",
+    "This guide shows how you can do this."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
     "## Setup\n",
     "\n",
-    "First we need to install the packages required"
+    "First, let's install the required packages"
    ]
   },
   {
@@ -60,7 +93,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {
@@ -68,7 +101,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Define SubGraph\n",
+    "## Define subgraph\n",
     "\n",
     "First, let's set up our subgraph. For this, we will create a simple graph that can get the weather for a specific city. We will compile this graph with a [breakpoint](https://langchain-ai.github.io/langgraph/how-tos/human_in_the_loop/breakpoints/) before the `weather_node`:"
    ]
@@ -121,7 +154,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Define Parent Graph\n",
+    "## Define parent graph\n",
     "\n",
     "We can now setup the overall graph. This graph will first route to the subgraph if it needs to get the weather, otherwise it will route to a normal LLM."
    ]
@@ -444,7 +477,7 @@
     "    if h.next == (\"model_node\",)\n",
     ")\n",
     "\n",
-    "# This pattern can be extended no matter how many levels deep - image model node was another subgraph in this case\n",
+    "# This pattern can be extended no matter how many levels deep\n",
     "# subsubgraph_stat_history = next(h for h in graph.get_state_history(subgraph_state_before_model_node.tasks[0].state) if h.next == ('my_subsubgraph_node',))"
    ]
   },
@@ -660,7 +693,9 @@
     "    print(update)\n",
     "# Graph execution should stop before the weather node\n",
     "print(\"interrupted!\")\n",
+    "\n",
     "state = graph.get_state(config, subgraphs=True)\n",
+    "\n",
     "# We update the state by passing in the message we want returned from the weather node, and make sure to use as_node\n",
     "graph.update_state(\n",
     "    state.tasks[0].state.config,\n",
@@ -669,6 +704,7 @@
     ")\n",
     "for update in graph.stream(None, config=config, stream_mode=\"updates\", subgraphs=True):\n",
     "    print(update)\n",
+    "\n",
     "print(graph.get_state(config).values[\"messages\"])"
    ]
   },
@@ -708,6 +744,7 @@
     "    print(update)\n",
     "# Graph execution should stop before the weather node\n",
     "print(\"interrupted!\")\n",
+    "\n",
     "# We update the state by passing in the message we want returned from the weather graph, making sure to use as_node\n",
     "# Note that we don't need to pass in the subgraph config, since we aren't updating the state inside the subgraph\n",
     "graph.update_state(\n",
@@ -717,6 +754,7 @@
     ")\n",
     "for update in graph.stream(None, config=config, stream_mode=\"updates\"):\n",
     "    print(update)\n",
+    "\n",
     "print(graph.get_state(config).values[\"messages\"])"
    ]
   },
@@ -947,6 +985,7 @@
     "    None, config=config, stream_mode=\"updates\", subgraphs=True\n",
     "):\n",
     "    print(update)\n",
+    "\n",
     "print(grandparent_graph.get_state(config).values[\"messages\"])"
    ]
   },
@@ -1002,7 +1041,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.9"
+   "version": "3.12.3"
   }
  },
  "nbformat": 4,

docs/docs/how-tos/tool-calling-errors.ipynb

@@ -76,7 +76,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/how-tos/tool-calling.ipynb

@@ -57,7 +57,7 @@
     "    <p style=\"padding-top: 5px;\">\n",
     "        Sign up for LangSmith to quickly spot issues and improve the performance of your LangGraph projects. LangSmith lets you use trace data to debug, test, and monitor your LLM apps built with LangGraph — read more about how to get started <a href=\"https://docs.smith.langchain.com\">here</a>. \n",
     "    </p>\n",
-    "</div>    "
+    "</div>"
    ]
   },
   {

docs/docs/index.md

@@ -3,7 +3,6 @@ hide_comments: true
 hide:
   - navigation
 title: Home
-
 ---
 
 {!README.md!}

docs/docs/reference/index.md

@@ -0,0 +1,10 @@
+---
+title: Reference
+description: API reference for LangGraph
+---
+
+# Reference
+
+Welcome to the LangGraph API reference! This reference provides detailed information about the LangGraph API, including classes, methods, and other components.
+
+If you are new to LangGraph, we recommend starting with the [Quick Start](../tutorials/introduction.ipynb) in the Tutorials section.

docs/docs/reference/remote_graph.md

@@ -0,0 +1,6 @@
+# RemoteGraph
+
+::: langgraph.pregel.remote
+    options:
+      members:
+        - RemoteGraph

docs/docs/troubleshooting/errors/GRAPH_RECURSION_LIMIT.md

@@ -0,0 +1,29 @@
+# GRAPH_RECURSION_LIMIT
+
+Your LangGraph [`StateGraph`](https://langchain-ai.github.io/langgraph/reference/graphs/#langgraph.graph.state.StateGraph) reached the maximum number of steps before hitting a stop condition.
+This is often due to an infinite loop caused by code like the example below:
+
+```python
+class State(TypedDict):
+    some_key: str
+
+builder = StateGraph(State)
+builder.add_node("a", ...)
+builder.add_node("b", ...)
+builder.add_edge("a", "b")
+builder.add_edge("b", "a")
+...
+
+graph = builder.compile()
+```
+
+However, complex graphs may hit the default limit naturally.
+
+## Troubleshooting
+
+- If you are not expecting your graph to go through many iterations, you likely have a cycle. Check your logic for infinite loops.
+- If you have a complex graph, you can pass in a higher `recursion_limit` value into your `config` object when invoking your graph like this:
+
+```python
+graph.invoke({...}, {"recursion_limit": 100})
+```

docs/docs/troubleshooting/errors/INVALID_CONCURRENT_GRAPH_UPDATE.md

@@ -0,0 +1,49 @@
+# INVALID_CONCURRENT_GRAPH_UPDATE
+
+A LangGraph [`StateGraph`](https://langchain-ai.github.io/langgraph/reference/graphs/#langgraph.graph.state.StateGraph) received concurrent updates to its state from multiple nodes to a state property that doesn't
+support it.
+
+One way this can occur is if you are using a [fanout](https://langchain-ai.github.io/langgraph/how-tos/map-reduce/)
+or other parallel execution in your graph and you have defined a graph like this:
+
+```python
+class State(TypedDict):
+    some_key: str
+
+def node(state: State):
+    return {"some_key": "some_string_value"}
+
+def other_node(state: State):
+    return {"some_key": "some_string_value"}
+
+
+builder = StateGraph(State)
+builder.add_node(node)
+builder.add_node(other_node)
+builder.add_edge(START, "node")
+builder.add_edge(START, "other_node")
+graph = builder.compile()
+```
+
+If a node in the above graph returns `{ "some_key": "some_string_value" }`, this will overwrite the state value for `"some_key"` with `"some_string_value"`.
+However, if multiple nodes in e.g. a fanout within a single step return values for `"some_key"`, the graph will throw this error because
+there is uncertainty around how to update the internal state.
+
+To get around this, you can define a reducer that combines multiple values:
+
+```python
+import operator
+from typing import Annotated
+
+class State(TypedDict):
+    # The operator.add reducer fn makes this append-only
+    some_key: Annotated[list, operator.add]
+```
+
+This will allow you to define logic that handles the same key returned from multiple nodes executed in parallel.
+
+## Troubleshooting
+
+The following may help resolve this error:
+
+- If your graph executes nodes in parallel, make sure you have defined relevant state keys with a reducer.

docs/docs/troubleshooting/errors/INVALID_GRAPH_NODE_RETURN_VALUE.md

@@ -0,0 +1,38 @@
+# INVALID_GRAPH_NODE_RETURN_VALUE
+
+A LangGraph [`StateGraph`](https://langchain-ai.github.io/langgraph/reference/graphs/#langgraph.graph.state.StateGraph)
+received a non-dict return type from a node. Here's an example:
+
+```python
+class State(TypedDict):
+    some_key: str
+
+def bad_node(state: State):
+    # Should return an dict with a value for "some_key", not a list
+    return ["whoops"]
+
+builder = StateGraph(State)
+builder.add_node(bad_node)
+...
+
+graph = builder.compile()
+```
+
+Invoking the above graph will result in an error like this:
+
+```python
+graph.invoke({ "some_key": "someval" });
+```
+
+```
+InvalidUpdateError: Expected dict, got ['whoops']
+For troubleshooting, visit: https://python.langchain.com/docs/troubleshooting/errors/INVALID_GRAPH_NODE_RETURN_VALUE
+```
+
+Nodes in your graph must return an dict containing one or more keys defined in your state.
+
+## Troubleshooting
+
+The following may help resolve this error:
+
+- If you have complex logic in your node, make sure all code paths return an appropriate dict for your defined state.