release(langgraph): 1.2.12

Bumps [soupsieve](https://github.com/facelessuser/soupsieve) from 2.8.4
to 2.9.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/facelessuser/soupsieve/releases">soupsieve's
releases</a>.</em></p>
<blockquote>
<h2>2.9</h2>
<ul>
<li><strong>NEW</strong>: Drop Python 3.9 support.</li>
<li><strong>NEW</strong>: Lazy compile selector patterns to improve
initial import speed.</li>
<li><strong>FIX</strong>: Correct
<code>:nth-child</code>/<code>:nth-of-type</code> (and
<code>-last-</code> variants) for <code>An+B</code> values whose
sequence steps onto
index 0 or onto the last child (e.g. <code>:nth-child(2n-2)</code>,
<code>:nth-child(n-1)</code>, <code>:nth-child(n+5)</code>), which
previously
matched the wrong elements or nothing at all (<a
href="https://github.com/gaoflow"><code>@​gaoflow</code></a>).</li>
<li><strong>FIX</strong>: More efficient CSS ID matching (<a
href="https://github.com/kaimandalic"><code>@​kaimandalic</code></a>).</li>
<li><strong>FIX</strong>: Fix inefficient trimming of comments and white
space (<a
href="https://github.com/kaimandalic"><code>@​kaimandalic</code></a>).</li>
</ul>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/8763f914472fc83652babda708bed5c8ef287004"><code>8763f91</code></a>
Format changelog message</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/cf198fcddc9230f06ed39f974eba0ce076b85cda"><code>cf198fc</code></a>
Fix inefficient trimming of comments and white space</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/ce44e4996e6632871c18cdd7a7fb641be8ef34ef"><code>ce44e49</code></a>
Merge commit from fork</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/751c57b2c7e978e206b94b7dba17f8e2af392e19"><code>751c57b</code></a>
Fix :nth-child/:nth-of-type matching for An+B index boundaries (<a
href="https://redirect.github.com/facelessuser/soupsieve/issues/297">#297</a>)</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/08e9ede4dcfafef860155319ef5eb9708e75d10b"><code>08e9ede</code></a>
Drop Python 3.9</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/d6e68303a6c3e0e410530939b92955ba24a07a81"><code>d6e6830</code></a>
Rework selector mapping</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/d2d1581fe275f89cb2e792589fed770aeb9e99b3"><code>d2d1581</code></a>
Utilize property for accessing lazy regular expression pattern</li>
<li><a
href="https://github.com/facelessuser/soupsieve/commit/b8701dec25c84a3910fd9a03222a3804fa119a1d"><code>b8701de</code></a>
Build patterns and regexes lazily in css_parser (<a
href="https://redirect.github.com/facelessuser/soupsieve/issues/296">#296</a>)</li>
<li>See full diff in <a
href="https://github.com/facelessuser/soupsieve/compare/2.8.4...2.9">compare
view</a></li>
</ul>
</details>
<br />


[![Dependabot compatibility
score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=soupsieve&package-manager=uv&previous-version=2.8.4&new-version=2.9)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)

Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the
[Security Alerts
page](https://github.com/langchain-ai/langgraph/network/alerts).

</details>

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

.github/workflows/deploy-redirects.yml

@@ -7,6 +7,10 @@ on:
     paths:
       - 'docs/**'
       - '.github/workflows/deploy-redirects.yml'
+  # llms.txt is fetched from docs.langchain.com at build time, so redeploy on a
+  # schedule to pick up docs changes that never touch this repo.
+  schedule:
+    - cron: '17 6 * * 1'
   workflow_dispatch:
 
 permissions:

docs/generate_redirects.py

@@ -12,13 +12,26 @@ which is SEO-friendly and treated similarly to 301 redirects by Google.
 To add new redirects, simply edit redirects.json and re-run this script.
 """
 
+import http.client
 import json
 import os
+import urllib.error
+import urllib.parse
+import urllib.request
 from pathlib import Path
 
 # Default fallback URL for any path not in the redirect map
 DEFAULT_REDIRECT = "https://docs.langchain.com/oss/python/langgraph/overview"
 
+# The docs site regenerates this index on every deploy, so fetching it here
+# keeps the published llms.txt from drifting. The URL is a hardcoded constant,
+# never built from input, and both it and the post-redirect URL are checked
+# against ALLOWED_LLMS_HOST before anything is read.
+CANONICAL_LLMS_URL = "https://docs.langchain.com/oss/python/langgraph/llms.txt"
+ALLOWED_LLMS_HOST = "docs.langchain.com"
+LLMS_FETCH_TIMEOUT = 30
+LLMS_MAX_BYTES = 1_000_000
+
 HTML_TEMPLATE = """<!doctype html>
 <html lang="en">
 <head>
@@ -75,6 +88,64 @@ CATCHALL_404_TEMPLATE = """<!doctype html>
 """
 
 
+def is_allowed_llms_url(url):
+    """Return True if url is HTTPS on the one host we accept content from."""
+    parsed = urllib.parse.urlsplit(url)
+    return parsed.scheme == "https" and parsed.hostname == ALLOWED_LLMS_HOST
+
+
+def fetch_canonical_llms_txt():
+    """Return the published LangGraph index, or None if it cannot be used.
+
+    Returning None leaves the caller on the committed docs/llms.txt, so a
+    docs.langchain.com outage degrades to a stale file rather than a broken
+    deploy or a published error page.
+    """
+    if not is_allowed_llms_url(CANONICAL_LLMS_URL):
+        print(f"Refusing to fetch {CANONICAL_LLMS_URL}: host not allowed")
+        return None
+
+    try:
+        with urllib.request.urlopen(  # noqa: S310 - constant, allowlisted URL
+            CANONICAL_LLMS_URL, timeout=LLMS_FETCH_TIMEOUT
+        ) as response:
+            # urlopen follows redirects, so re-check where it actually landed.
+            if not is_allowed_llms_url(response.url):
+                print(f"Refusing {CANONICAL_LLMS_URL}: redirected to {response.url}")
+                return None
+            body = response.read(LLMS_MAX_BYTES + 1)
+    # A connection dropped mid-body raises http.client.IncompleteRead, which
+    # descends from HTTPException rather than OSError, so catching only the
+    # urllib and OS errors would let it escape and fail the whole deploy.
+    except (
+        urllib.error.URLError,
+        http.client.HTTPException,
+        TimeoutError,
+        OSError,
+    ) as exc:
+        print(f"Could not fetch {CANONICAL_LLMS_URL}: {type(exc).__name__}: {exc}")
+        return None
+
+    if len(body) > LLMS_MAX_BYTES:
+        print(f"Refusing {CANONICAL_LLMS_URL}: larger than {LLMS_MAX_BYTES} bytes")
+        return None
+
+    try:
+        text = body.decode("utf-8")
+    except UnicodeDecodeError as exc:
+        print(f"Refusing {CANONICAL_LLMS_URL}: not valid UTF-8: {exc}")
+        return None
+
+    # An index opens with a markdown heading and links to the docs site. A
+    # body that does not is an error page or a truncated response, not content
+    # worth publishing.
+    if not text.startswith("# ") or f"https://{ALLOWED_LLMS_HOST}/" not in text:
+        print(f"Refusing {CANONICAL_LLMS_URL}: does not look like an llms.txt index")
+        return None
+
+    return text
+
+
 def generate_redirects():
     script_dir = Path(__file__).parent
     output_dir = script_dir / "_site"
@@ -126,14 +197,19 @@ def generate_redirects():
     catchall_404.write_text(CATCHALL_404_TEMPLATE.format(default_url=DEFAULT_REDIRECT))
     print(f"Created: {catchall_404}")
 
-    # Copy static files (like llms.txt) that can't be redirected via HTML
-    static_files = ["llms.txt"]
-    for static_file in static_files:
-        src = script_dir / static_file
+    # llms.txt can't be redirected via HTML, so publish the docs site's own
+    # generated index. The committed copy is only a fallback.
+    llms_txt = fetch_canonical_llms_txt()
+    if llms_txt is not None:
+        (output_dir / "llms.txt").write_text(llms_txt)
+        print(f"Fetched: {output_dir / 'llms.txt'} (from {CANONICAL_LLMS_URL})")
+    else:
+        src = script_dir / "llms.txt"
         if src.exists():
-            dst = output_dir / static_file
-            dst.write_text(src.read_text())
-            print(f"Copied: {dst}")
+            (output_dir / "llms.txt").write_text(src.read_text())
+            print(f"Copied: {output_dir / 'llms.txt'} (fallback, may be stale)")
+        else:
+            print("No llms.txt fetched and no committed fallback; skipping")
 
     print(f"\nGenerated {len(redirects)} redirect files in {output_dir}")
 

docs/llms.txt

@@ -1,35 +1,49 @@
-# LangGraph
+# Docs by LangChain: LangGraph (Python)
 
-LangGraph documentation has moved to docs.langchain.com.
+> Markdown index of the LangGraph (Python) documentation.
 
-## Overview
+## LangGraph (Python)
 
-- [LangGraph Overview](https://docs.langchain.com/oss/python/langgraph/overview): Introduction to LangGraph, a library for building stateful, multi-actor applications with LLMs.
-- [Why LangGraph?](https://docs.langchain.com/oss/python/langgraph/why-langgraph): Motivation for LangGraph and its key features.
-
-## Core Concepts
-
-- [Graph API](https://docs.langchain.com/oss/python/langgraph/graph-api): Learn how to define state, create nodes, and connect them with edges.
-- [Streaming](https://docs.langchain.com/oss/python/langgraph/streaming): Stream outputs from your graph for better UX.
-- [Persistence](https://docs.langchain.com/oss/python/langgraph/persistence): Add memory and checkpointing to your graphs.
-- [Add Memory](https://docs.langchain.com/oss/python/langgraph/add-memory): Implement short-term and long-term memory.
-- [Workflows & Agents](https://docs.langchain.com/oss/python/langgraph/workflows-agents): Build agents and workflows with LangGraph.
-
-## How-To Guides
-
-- [Use Subgraphs](https://docs.langchain.com/oss/python/langgraph/use-subgraphs): Compose graphs using subgraphs.
-- [Observability](https://docs.langchain.com/oss/python/langgraph/observability): Add tracing and debugging to your graphs.
-- [Common Errors](https://docs.langchain.com/oss/python/langgraph/common-errors): Troubleshoot common LangGraph errors.
-
-## Tutorials
-
-- [Agentic RAG](https://docs.langchain.com/oss/python/langgraph/agentic-rag): Build an agentic RAG system with LangGraph.
-- [SQL Agent](https://docs.langchain.com/oss/python/langgraph/sql-agent): Create a SQL agent with LangGraph.
-
-## Reference
-
-- [API Reference](https://reference.langchain.com/python/langgraph/): Complete API documentation for LangGraph.
-
-## LangGraph Platform
-
-For deploying LangGraph applications in production, see the [LangSmith documentation](https://docs.langchain.com/langsmith/agent-server).
+- [Memory](https://docs.langchain.com/oss/python/langgraph/add-memory.md)
+- [Build a custom RAG agent with LangGraph](https://docs.langchain.com/oss/python/langgraph/agentic-rag.md)
+- [Application structure](https://docs.langchain.com/oss/python/langgraph/application-structure.md)
+- [Backward compatibility](https://docs.langchain.com/oss/python/langgraph/backward-compatibility.md)
+- [Case studies](https://docs.langchain.com/oss/python/langgraph/case-studies.md)
+- [Changelog](https://docs.langchain.com/oss/python/langgraph/changelog-js.md)
+- [Changelog](https://docs.langchain.com/oss/python/langgraph/changelog-py.md)
+- [Checkpointers](https://docs.langchain.com/oss/python/langgraph/checkpointers.md)
+- [Choosing between Graph and Functional APIs](https://docs.langchain.com/oss/python/langgraph/choosing-apis.md)
+- [Deployment](https://docs.langchain.com/oss/python/langgraph/deploy.md)
+- [GRAPH_RECURSION_LIMIT](https://docs.langchain.com/oss/python/langgraph/errors/GRAPH_RECURSION_LIMIT.md)
+- [INVALID_CHAT_HISTORY](https://docs.langchain.com/oss/python/langgraph/errors/INVALID_CHAT_HISTORY.md)
+- [INVALID_CONCURRENT_GRAPH_UPDATE](https://docs.langchain.com/oss/python/langgraph/errors/INVALID_CONCURRENT_GRAPH_UPDATE.md)
+- [INVALID_GRAPH_NODE_RETURN_VALUE](https://docs.langchain.com/oss/python/langgraph/errors/INVALID_GRAPH_NODE_RETURN_VALUE.md)
+- [MISSING_CHECKPOINTER](https://docs.langchain.com/oss/python/langgraph/errors/MISSING_CHECKPOINTER.md)
+- [MULTIPLE_SUBGRAPHS](https://docs.langchain.com/oss/python/langgraph/errors/MULTIPLE_SUBGRAPHS.md)
+- [Event streaming](https://docs.langchain.com/oss/python/langgraph/event-streaming.md)
+- [Fault tolerance](https://docs.langchain.com/oss/python/langgraph/fault-tolerance.md)
+- [Custom stream channels](https://docs.langchain.com/oss/python/langgraph/frontend/custom-stream-channels.md)
+- [Graph execution](https://docs.langchain.com/oss/python/langgraph/frontend/graph-execution.md)
+- [Overview](https://docs.langchain.com/oss/python/langgraph/frontend/overview.md)
+- [Functional API overview](https://docs.langchain.com/oss/python/langgraph/functional-api.md)
+- [Graph API overview](https://docs.langchain.com/oss/python/langgraph/graph-api.md)
+- [Install LangGraph](https://docs.langchain.com/oss/python/langgraph/install.md)
+- [Interrupts](https://docs.langchain.com/oss/python/langgraph/interrupts.md)
+- [Run a local server](https://docs.langchain.com/oss/python/langgraph/local-server.md)
+- [LangSmith Observability](https://docs.langchain.com/oss/python/langgraph/observability.md)
+- [LangGraph overview](https://docs.langchain.com/oss/python/langgraph/overview.md)
+- [Persistence](https://docs.langchain.com/oss/python/langgraph/persistence.md)
+- [LangGraph runtime](https://docs.langchain.com/oss/python/langgraph/pregel.md)
+- [Quickstart](https://docs.langchain.com/oss/python/langgraph/quickstart.md)
+- [Build a custom SQL agent](https://docs.langchain.com/oss/python/langgraph/sql-agent.md)
+- [Stores](https://docs.langchain.com/oss/python/langgraph/stores.md)
+- [Streaming](https://docs.langchain.com/oss/python/langgraph/streaming.md)
+- [LangSmith Studio](https://docs.langchain.com/oss/python/langgraph/studio.md)
+- [Test](https://docs.langchain.com/oss/python/langgraph/test.md)
+- [Thinking in LangGraph](https://docs.langchain.com/oss/python/langgraph/thinking-in-langgraph.md)
+- [Agent Chat UI](https://docs.langchain.com/oss/python/langgraph/ui.md)
+- [Use the functional API](https://docs.langchain.com/oss/python/langgraph/use-functional-api.md)
+- [Use the graph API](https://docs.langchain.com/oss/python/langgraph/use-graph-api.md)
+- [Subgraphs](https://docs.langchain.com/oss/python/langgraph/use-subgraphs.md)
+- [Use time-travel](https://docs.langchain.com/oss/python/langgraph/use-time-travel.md)
+- [Workflows and agents](https://docs.langchain.com/oss/python/langgraph/workflows-agents.md)

libs/cli/langgraph_cli/deploy.py

@@ -51,7 +51,6 @@ RESERVED_ENV_VARS = frozenset(
         "LANGGRAPH_AUTH_TYPE",
         "LANGSMITH_AUTH_ENDPOINT",
         "LANGSMITH_TENANT_ID",
-        "LANGSMITH_WORKSPACE_ID",
         "LANGSMITH_AUTH_VERIFY_TENANT_ID",
         "LANGSMITH_HOST_PROJECT_ID",
         "LANGSMITH_HOST_PROJECT_NAME",
@@ -1230,24 +1229,6 @@ def _run_remote_build(
 # ---------------------------------------------------------------------------
 
 
-def _get_tenant_id(env_vars: dict[str, str]) -> str | None:
-    """Get the tenant ID from LANGSMITH_TENANT_ID or LANGSMITH_WORKSPACE_ID."""
-    tenant_id = env_vars.get("LANGSMITH_TENANT_ID") or os.environ.get(
-        "LANGSMITH_TENANT_ID"
-    )
-    fallback_tenant_id = env_vars.get("LANGSMITH_WORKSPACE_ID") or os.environ.get(
-        "LANGSMITH_WORKSPACE_ID"
-    )
-    if tenant_id and fallback_tenant_id:
-        raise click.UsageError(
-            "LANGSMITH_TENANT_ID and LANGSMITH_WORKSPACE_ID cannot both be set. "
-            "Set only one."
-        )
-    if tenant_id:
-        return tenant_id
-    return fallback_tenant_id or None
-
-
 def _create_host_backend_client(
     host_url: str | None,
     api_key: str | None,
@@ -1255,7 +1236,6 @@ def _create_host_backend_client(
 ) -> HostBackendClient:
     if env_vars is None:
         env_vars = _parse_env_from_config({}, pathlib.Path.cwd() / DEFAULT_CONFIG)
-    tenant_id = _get_tenant_id(env_vars)
     resolved_api_key = api_key
     if not resolved_api_key:
         for key_name in _API_KEY_ENV_NAMES:
@@ -1278,6 +1258,9 @@ def _create_host_backend_client(
             fg="yellow",
         )
         resolved_api_key = click.prompt("Enter LangSmith API key", hide_input=True)
+    tenant_id = env_vars.get("LANGSMITH_TENANT_ID") or os.environ.get(
+        "LANGSMITH_TENANT_ID"
+    )
     return HostBackendClient(host_url, resolved_api_key, tenant_id=tenant_id)
 
 

libs/cli/tests/unit_tests/test_deploy_helpers.py

@@ -21,7 +21,6 @@ from langgraph_cli.deploy import (
     _parse_env_from_config,
     _resolve_env_path,
     _resolve_pushed_image_digest,
-    _secrets_from_env,
     _smith_dashboard_base_url,
     _validate_prebuilt_image,
     normalize_image_tag,
@@ -541,63 +540,6 @@ class TestCreateHostBackendClientNoInput:
         assert client is not None
 
 
-@pytest.mark.parametrize("source", ["config", "shell"])
-@pytest.mark.parametrize("name", ["LANGSMITH_TENANT_ID", "LANGSMITH_WORKSPACE_ID"])
-def test_workspace_id_alias(monkeypatch, source, name):
-    monkeypatch.delenv("LANGSMITH_WORKSPACE_ID", raising=False)
-    monkeypatch.delenv("LANGSMITH_TENANT_ID", raising=False)
-    env_vars = {}
-    if source == "config":
-        env_vars[name] = "workspace"
-    else:
-        monkeypatch.setenv(name, "workspace")
-
-    def handler(request):
-        assert request.headers["X-Tenant-ID"] == "workspace"
-        return httpx.Response(200, json={"resources": []})
-
-    monkeypatch.setattr(
-        httpx, "HTTPTransport", lambda **kwargs: httpx.MockTransport(handler)
-    )
-    client = _create_host_backend_client(
-        "https://api.example.com", "test-key", env_vars
-    )
-    try:
-        client.list_deployments()
-    finally:
-        client._client.close()
-
-
-@pytest.mark.parametrize("tenant_source", ["config", "shell"])
-@pytest.mark.parametrize("workspace_source", ["config", "shell"])
-@pytest.mark.parametrize("workspace_id", ["tenant-id", "workspace-id"])
-def test_rejects_both_workspace_names(
-    monkeypatch, tenant_source, workspace_source, workspace_id
-):
-    env_vars = {}
-    for name, source, value in [
-        ("LANGSMITH_TENANT_ID", tenant_source, "tenant-id"),
-        ("LANGSMITH_WORKSPACE_ID", workspace_source, workspace_id),
-    ]:
-        monkeypatch.delenv(name, raising=False)
-        if source == "config":
-            env_vars[name] = value
-        else:
-            monkeypatch.setenv(name, value)
-
-    with pytest.raises(
-        click.UsageError,
-        match="LANGSMITH_TENANT_ID and LANGSMITH_WORKSPACE_ID cannot both be set",
-    ):
-        _create_host_backend_client("https://api.example.com", "test-key", env_vars)
-
-
-def test_workspace_id_is_not_uploaded_as_secret():
-    assert _secrets_from_env(
-        {"LANGSMITH_WORKSPACE_ID": "workspace", "APP_SETTING": "value"}
-    ) == [{"name": "APP_SETTING", "value": "value"}]
-
-
 class TestSmithDashboardBaseUrl:
     def test_none_returns_default(self):
         assert _smith_dashboard_base_url(None) == "https://smith.langchain.com"

libs/langgraph/langgraph/types.py

@@ -12,15 +12,23 @@ from typing import (
     Generic,
     Literal,
     NamedTuple,
-    TypeVar,
     final,
+    overload,
 )
 from warnings import warn
 
 from langchain_core.messages import AnyMessage
 from langchain_core.runnables import Runnable, RunnableConfig
 from langgraph.checkpoint.base import BaseCheckpointSaver, CheckpointMetadata
-from typing_extensions import NotRequired, TypeAliasType, TypedDict, Unpack, deprecated
+from pydantic import TypeAdapter
+from typing_extensions import (
+    NotRequired,
+    TypeAliasType,
+    TypedDict,
+    TypeVar,
+    Unpack,
+    deprecated,
+)
 from xxhash import xxh3_128_hexdigest
 
 from langgraph._internal._cache import default_cache_key
@@ -36,6 +44,7 @@ from langgraph.warnings import LangGraphDeprecatedSinceV10, LangGraphDeprecatedS
 # when used in standalone type aliases.
 StateT = TypeVar("StateT")
 OutputT = TypeVar("OutputT")
+ResponseT = TypeVar("ResponseT", default=Any)
 
 if TYPE_CHECKING:
     from langgraph.pregel.protocol import PregelProtocol
@@ -572,7 +581,7 @@ _DEFAULT_INTERRUPT_ID = "placeholder-id"
 
 @final
 @dataclass(init=False, slots=True)
-class Interrupt:
+class Interrupt(Generic[ResponseT]):
     """Information about an interrupt that occurred in a node.
 
     !!! version-added "Added in version 0.2.24"
@@ -596,13 +605,22 @@ class Interrupt:
     id: str
     """The ID of the interrupt. Can be used to resume the interrupt directly."""
 
+    response_schema: type[ResponseT] | dict[str, Any] | None = None
+    """Schema for the value expected when resuming this interrupt, if the graph provided one.
+
+    A surfaced interrupt carries JSON Schema (a `dict`); `type[ResponseT]` records the
+    Python type at construction so `Interrupt[Decision]` is meaningful to type checkers."""
+
     def __init__(
         self,
         value: Any,
         id: str = _DEFAULT_INTERRUPT_ID,
+        *,
+        response_schema: type[ResponseT] | dict[str, Any] | None = None,
         **deprecated_kwargs: Unpack[DeprecatedKwargs],
     ) -> None:
         self.value = value
+        self.response_schema = response_schema
 
         if (
             (ns := deprecated_kwargs.get("ns", MISSING)) is not MISSING
@@ -614,8 +632,18 @@ class Interrupt:
             self.id = id
 
     @classmethod
-    def from_ns(cls, value: Any, ns: str) -> Interrupt:
-        return cls(value=value, id=xxh3_128_hexdigest(ns.encode()))
+    def from_ns(
+        cls,
+        value: Any,
+        ns: str,
+        *,
+        response_schema: type[ResponseT] | dict[str, Any] | None = None,
+    ) -> Interrupt[ResponseT]:
+        return cls(
+            value=value,
+            id=xxh3_128_hexdigest(ns.encode()),
+            response_schema=response_schema,
+        )
 
     @property
     @deprecated("`interrupt_id` is deprecated. Use `id` instead.", category=None)
@@ -848,7 +876,17 @@ class Command(Generic[N], ToolOutputMixin):
     PARENT: ClassVar[Literal["__parent__"]] = "__parent__"
 
 
-def interrupt(value: Any) -> Any:
+@overload
+def interrupt(value: Any, *, response_schema: type[ResponseT]) -> ResponseT: ...
+
+
+@overload
+def interrupt(value: Any, *, response_schema: dict[str, Any] | None = None) -> Any: ...
+
+
+def interrupt(
+    value: Any, *, response_schema: dict[str, Any] | type | None = None
+) -> Any:
     """Interrupt the graph with a resumable exception from within a node.
 
     The `interrupt` function enables human-in-the-loop workflows by pausing graph
@@ -918,7 +956,7 @@ def interrupt(value: Any) -> Any:
         for chunk in graph.stream({\"foo\": \"abc\"}, config):
             print(chunk)
 
-        # > {'__interrupt__': (Interrupt(value='what is your age?', id='45fda8478b2ef754419799e10992af06'),)}
+        # > {'__interrupt__': (Interrupt(value='what is your age?', id='45fda8478b2ef754419799e10992af06', response_schema=None),)}
 
         command = Command(resume=\"some input from a human!!!\")
 
@@ -931,12 +969,20 @@ def interrupt(value: Any) -> Any:
 
     Args:
         value: The value to surface to the client when the graph is interrupted.
+        response_schema: Optional schema for the value expected on resume, surfaced
+            to clients so they can render a typed input form. Accepts a JSON Schema
+            `dict` (used as-is, resume values are not validated), or a Pydantic model
+            class, `TypedDict`, or dataclass, which are converted to JSON Schema for
+            clients and used to validate the resume value; the validated object is
+            what `interrupt` returns.
 
     Returns:
-        Any: On subsequent invocations within the same node (same task to be precise), returns the value provided during the first invocation
+        Any: On subsequent invocations within the same node (same task to be precise), returns the value provided during the first invocation,
+            validated against `response_schema` when one that supports validation was given.
 
     Raises:
         GraphInterrupt: On the first invocation within the node, halts execution and surfaces the provided value to the client.
+        pydantic.ValidationError: When a resume value does not match a Pydantic model, `TypedDict`, or dataclass `response_schema`.
     """
     from langgraph._internal._constants import (
         CONFIG_KEY_CHECKPOINT_NS,
@@ -948,27 +994,36 @@ def interrupt(value: Any) -> Any:
     from langgraph.errors import GraphInterrupt
 
     conf = get_config()["configurable"]
+    adapter = (
+        None
+        if response_schema is None or isinstance(response_schema, dict)
+        else TypeAdapter(response_schema)
+    )
     # track interrupt index
     scratchpad = conf[CONFIG_KEY_SCRATCHPAD]
     idx = scratchpad.interrupt_counter()
     # find previous resume values
     if scratchpad.resume:
         if idx < len(scratchpad.resume):
-            conf[CONFIG_KEY_SEND]([(RESUME, scratchpad.resume)])
-            return scratchpad.resume[idx]
+            v = scratchpad.resume[idx]
+            validated = adapter.validate_python(v) if adapter else v
+            conf[CONFIG_KEY_SEND]([(RESUME, scratchpad.resume[: idx + 1])])
+            return validated
     # find current resume value
     v = scratchpad.get_null_resume(True)
     if v is not None:
         assert len(scratchpad.resume) == idx, (scratchpad.resume, idx)
+        validated = adapter.validate_python(v) if adapter else v
         scratchpad.resume.append(v)
         conf[CONFIG_KEY_SEND]([(RESUME, scratchpad.resume)])
-        return v
+        return validated
     # no resume value found
     raise GraphInterrupt(
         (
             Interrupt.from_ns(
                 value=value,
                 ns=conf[CONFIG_KEY_CHECKPOINT_NS],
+                response_schema=adapter.json_schema() if adapter else response_schema,
             ),
         )
     )

libs/langgraph/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "langgraph"
-version = "1.2.11"
+version = "1.2.12"
 description = "Building stateful, multi-actor applications with LLMs"
 authors = []
 requires-python = ">=3.10"

libs/langgraph/tests/test_interruption.py

@@ -1,9 +1,14 @@
+from dataclasses import dataclass
+from typing import Any
+
 import pytest
 from langgraph.checkpoint.base import BaseCheckpointSaver
+from pydantic import BaseModel, ValidationError
 from typing_extensions import TypedDict
 
 from langgraph.graph import END, START, StateGraph
-from langgraph.types import Durability
+from langgraph.types import Command, Durability, Interrupt, interrupt
+from tests.any_str import AnyStr
 
 pytestmark = pytest.mark.anyio
 
@@ -90,3 +95,150 @@ async def test_interruption_without_state_updates_async(
     assert (await graph.aget_state(thread)).next == ()
     n_checkpoints = len([c async for c in graph.aget_state_history(thread)])
     assert n_checkpoints == (5 if durability != "exit" else 3)
+
+
+class Decision(BaseModel):
+    approved: bool
+    note: str | None = None
+
+
+class DecisionDict(TypedDict):
+    approved: bool
+
+
+@dataclass
+class DecisionData:
+    approved: bool
+
+
+RAW_SCHEMA = {"type": "object", "properties": {"approved": {"type": "boolean"}}}
+
+
+@pytest.mark.parametrize(
+    ("response_schema", "expected_schema", "expected_answer"),
+    [
+        (None, None, {"approved": True, "extra": 1}),
+        (RAW_SCHEMA, RAW_SCHEMA, {"approved": True, "extra": 1}),
+        (Decision, Decision.model_json_schema(), Decision(approved=True)),
+        (
+            DecisionDict,
+            {
+                "properties": {"approved": {"title": "Approved", "type": "boolean"}},
+                "required": ["approved"],
+                "title": "DecisionDict",
+                "type": "object",
+            },
+            {"approved": True},
+        ),
+        (
+            DecisionData,
+            {
+                "properties": {"approved": {"title": "Approved", "type": "boolean"}},
+                "required": ["approved"],
+                "title": "DecisionData",
+                "type": "object",
+            },
+            DecisionData(approved=True),
+        ),
+    ],
+    ids=["none", "raw_dict", "pydantic", "typeddict", "dataclass"],
+)
+def test_interrupt_response_schema(
+    sync_checkpointer: BaseCheckpointSaver,
+    response_schema: Any,
+    expected_schema: dict[str, Any] | None,
+    expected_answer: Any,
+) -> None:
+    class State(TypedDict):
+        answer: Any
+
+    def node(state: State) -> State:
+        return {
+            "answer": interrupt(
+                {"question": "approve?"}, response_schema=response_schema
+            )
+        }
+
+    graph = (
+        StateGraph(State)
+        .add_node("node", node)
+        .add_edge(START, "node")
+        .compile(checkpointer=sync_checkpointer)
+    )
+    config = {"configurable": {"thread_id": "1"}}
+    expected = Interrupt(
+        value={"question": "approve?"}, id=AnyStr(), response_schema=expected_schema
+    )
+
+    assert list(graph.stream({"answer": None}, config)) == [
+        {"__interrupt__": (expected,)}
+    ]
+    assert graph.get_state(config).tasks[0].interrupts == (expected,)
+    assert graph.invoke(Command(resume={"approved": True, "extra": 1}), config) == {
+        "answer": expected_answer
+    }
+
+
+@pytest.mark.parametrize("resume_style", ["null", "map"])
+def test_interrupt_response_schema_rejects_invalid_resume(
+    sync_checkpointer: BaseCheckpointSaver, resume_style: str
+) -> None:
+    class State(TypedDict):
+        answer: Any
+
+    def node(state: State) -> State:
+        return {"answer": interrupt("approve?", response_schema=Decision)}
+
+    graph = (
+        StateGraph(State)
+        .add_node("node", node)
+        .add_edge(START, "node")
+        .compile(checkpointer=sync_checkpointer)
+    )
+    config = {"configurable": {"thread_id": "1"}}
+    graph.invoke({"answer": None}, config)
+    [pending] = graph.get_state(config).tasks[0].interrupts
+
+    def resume(value: dict[str, Any]) -> Command:
+        return Command(resume=value if resume_style == "null" else {pending.id: value})
+
+    with pytest.raises(ValidationError, match="approved"):
+        graph.invoke(resume({"approved": "nope"}), config)
+
+    assert graph.invoke(resume({"approved": False}), config) == {
+        "answer": Decision(approved=False)
+    }
+
+
+@pytest.mark.parametrize("resume_style", ["null", "id_map"])
+def test_interrupt_response_schema_invalid_resume_after_earlier_interrupt(
+    sync_checkpointer: BaseCheckpointSaver, resume_style: str
+) -> None:
+    class State(TypedDict):
+        answer: Any
+
+    def node(state: State) -> State:
+        first = interrupt("first")
+        second = interrupt("approve?", response_schema=Decision)
+        return {"answer": [first, second]}
+
+    graph = (
+        StateGraph(State)
+        .add_node("node", node)
+        .add_edge(START, "node")
+        .compile(checkpointer=sync_checkpointer)
+    )
+    config = {"configurable": {"thread_id": "1"}}
+    graph.invoke({"answer": None}, config)
+    graph.invoke(Command(resume="ok"), config)
+    [pending] = graph.get_state(config).tasks[0].interrupts
+
+    def resume(value: dict[str, Any]) -> Command:
+        return Command(resume=value if resume_style == "null" else {pending.id: value})
+
+    with pytest.raises(ValidationError, match="approved"):
+        graph.invoke(resume({"approved": "nope"}), config)
+
+    assert graph.invoke(resume({"approved": True}), config) == {
+        "answer": ["ok", Decision(approved=True)]
+    }

libs/langgraph/tests/test_pregel.py

@@ -5583,6 +5583,7 @@ def test_falsy_return_from_task(sync_checkpointer: BaseCheckpointSaver):
                 "interrupts": [
                     {
                         "id": AnyStr(),
+                        "response_schema": None,
                         "value": "test",
                     },
                 ],
@@ -5627,6 +5628,7 @@ def test_falsy_return_from_task(sync_checkpointer: BaseCheckpointSaver):
                         "interrupts": (
                             {
                                 "id": AnyStr(),
+                                "response_schema": None,
                                 "value": "test",
                             },
                         ),

libs/langgraph/uv.lock

@@ -1437,7 +1437,7 @@ wheels = [
 
 [[package]]
 name = "langgraph"
-version = "1.2.11"
+version = "1.2.12"
 source = { editable = "." }
 dependencies = [
     { name = "langchain-core" },
@@ -3404,11 +3404,11 @@ wheels = [
 
 [[package]]
 name = "soupsieve"
-version = "2.8.4"
+version = "2.9"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/47/2c/0a5f6f8ee0d5589e48c7640213ed5175d52cf540a06725b628cc1a45d6ce/soupsieve-2.8.4.tar.gz", hash = "sha256:e121fd02e975c695e4e9e8774a5ee35d74714b59307868dcc5319ad2d9e3328e", size = 121110, upload-time = "2026-05-24T13:55:57.154Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/80/f1/93422647dd7e461f23d254e6b2bfa687a85b53aeb4903fcdbb74474d4584/soupsieve-2.9.tar.gz", hash = "sha256:acee8417325c5653e1377dc31eccad59eb82cbc65942afe6174c53b3aaad63fc", size = 122122, upload-time = "2026-07-19T01:35:18.425Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/5e/f5/0c41cb68dcae6b7de4fac4188a3a9589e21fb31df21ea3a2e888db95e6c9/soupsieve-2.8.4-py3-none-any.whl", hash = "sha256:e7e6b0769c8f51ed59acab6e994b00621096cfb1c640a7509295987388fbaf65", size = 37304, upload-time = "2026-05-24T13:55:55.406Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/d6/3185ab5ad1280319b31986898f3206dd7227cd75e293d4dba2a5e6bf27a0/soupsieve-2.9-py3-none-any.whl", hash = "sha256:a2b2c76d67df2382d245409fd71e321a571717e58463efa32ace87dcadac2c12", size = 37387, upload-time = "2026-07-19T01:35:17.106Z" },
 ]
 
 [[package]]

libs/prebuilt/uv.lock

@@ -286,7 +286,7 @@ wheels = [
 
 [[package]]
 name = "langgraph"
-version = "1.2.11"
+version = "1.2.12"
 source = { editable = "../langgraph" }
 dependencies = [
     { name = "langchain-core" },

libs/sdk-py/langgraph_sdk/schema.py

@@ -295,6 +295,8 @@ class Interrupt(TypedDict):
     """The value associated with the interrupt."""
     id: str
     """The ID of the interrupt. Can be used to resume the interrupt."""
+    response_schema: NotRequired[dict[str, Any]]
+    """JSON Schema for the value expected when resuming this interrupt, if the graph provided one."""
 
 
 class Thread(TypedDict):

libs/sdk-py/uv.lock

@@ -299,7 +299,7 @@ wheels = [
 
 [[package]]
 name = "langgraph"
-version = "1.2.11"
+version = "1.2.12"
 source = { editable = "../langgraph" }
 dependencies = [
     { name = "langchain-core" },