Commit Graph
62 Commits
Author SHA1 Message Date
Sydney RunkleandGitHub 0171e9a323 fix(langgraph): remove deprecated output usage in favor of output_schema (#5095)
use output_schema
2025-06-13 12:34:39 -04:00
Sydney RunkleandGitHub 3ad061f0d7 serialize/deserialize pandas with pickle fallback (#5057) 2025-06-12 15:14:00 -04:00
Emmanuel FerdmanandGitHub d719438307 fix: throw exception on multiple injections (#5033)
Throw exception on for multiple injections

Signed-off-by: Emmanuel Ferdman <emmanuelferdman@gmail.com>
2025-06-10 16:54:01 -04:00
Nuno CamposandGitHub 2e33c520a5 Support numpy array serialization in JsonPlusSerializer (#5035)
* Handle numpy Fortran arrays

* Lint

* Lint

* Lint
2025-06-10 01:17:28 +00:00
Nuno CamposandGitHub 67b1dc602e Update ormsgpack (#5034)
* Update ormsgpack

- Now supports bytearray/memoryview passthrough

* Lint
2025-06-10 00:30:58 +00:00
Sydney RunkleandGitHub b735452153 deprecate input and output in favor of input_schema and output_schema (#4983) 2025-06-06 19:44:56 -04:00
a5e6223569 Remove MessageGraph (#4875)
Remove MessageGraph

Co-authored-by: Nuno Campos <nuno@langchain.dev>
2025-05-29 21:50:23 -04:00
Nuno CamposandSydney Runkle afb83d2201 Remove non-state Graph
- Not used in any examples/docs
2025-05-29 21:38:58 -04:00
Nuno CamposandGitHub 76661f4c9c Remove Checkpoint.writes (#4822) 2025-05-28 14:04:45 -07:00
Nuno CamposandGitHub c32677c92d Remove old checkpoint test fixtures (#4814) 2025-05-28 14:03:12 -07:00
Sydney RunkleandGitHub 5602c29668 prebuilt: release 0.2.2 (#4851)
bump prebuilt
2025-05-28 13:33:39 +00:00
Sydney RunkleandGitHub 19ca6b416b fix: post_model_hook should inject state + store into tool calls if necessary (#4842) 2025-05-27 17:42:20 +00:00
Nuno Campos 936176eb21 Remove Checkpoint.writes
- This has been superseded by saving the individual writes of each task through put_writes()
- Removing this speeds up checkpoint operations as it was duplicating data saved elsewhere already
2025-05-25 19:06:46 -07:00
Nuno Campos bc17c3522b Remove old checkpoint test fixtures
- Now all tests fully migrated to more recent sync_checkpointer and async_checkpointer fixtures for parametrising on checkpointer
- Use sync/async_store fixtures where tests used only in memory store
- Remove unused "should snapshot" check for older versions of langchain core no longer tested against
2025-05-24 14:27:13 -07:00
Nuno Campos 0139e11ae5 0.4.7 2025-05-23 16:51:48 -07:00
Nuno Campos 9170f636d0 Lock 2025-05-23 16:14:15 -07:00
Sydney RunkleandGitHub a9c87ed8b6 prebuilt: release 0.2.1 (#4801)
lockfile and version updates
2025-05-23 18:07:24 +00:00
Sydney RunkleandGitHub 837fe59e24 prebuilt: support provider builtin tools in create_react_agent (#4800) 2025-05-23 13:58:53 -04:00
Sydney RunkleandGitHub aff3be03dc prebuilt: release 0.2.0 (#4793) 2025-05-22 17:54:58 +00:00
Sydney RunkleandGitHub 87fa661ac0 temp: remove ToolInterruptNode code (#4792)
remove tool node specific code for post_model_hook release
2025-05-22 13:28:59 -04:00
Tat Dat Duong 461afe0aa6 Again? 2025-05-22 00:12:40 +02:00
Sydney RunkleandGitHub 06607e08ea prebuilts hitl: fix branching logic + add structural snapshot tests (#4767) 2025-05-21 09:29:59 -04:00
Sydney RunkleandGitHub 7bd8616b1e feature: Implement post_model_hook and HumanInterruptNode (#4583) 2025-05-19 14:07:42 -04:00
William FHandGitHub 95f92069a7 sqlite: Add test for search with list filters (#4747) 2025-05-18 23:50:27 -07:00
William FHandGitHub 6b28319796 sqlite: update list_namespaces with max_depth (#4746)
sqlite: update on conflict
2025-05-18 23:27:06 -07:00
William FHandGitHub 725dd40fa7 Release sqlite store (#4737) 2025-05-17 21:49:03 -07:00
William FHandGitHub 025b634d98 SqliteStore (#3608) 2025-05-17 21:40:19 -07:00
Sydney RunkleandGitHub 228a08b966 ci: migrate to uv! (#4698)
* Migrate to `uv`
* Format `pyproject.toml` files properly
* Remove upper bounds on dependencies, and bounds on dev dependencies
(we should be using latest)
* Move to hatch for packaing

In the future we should:
* Set up dependabot / automate lockfile updates and tests
* Add tests for min compatible versions (I'll do this right after merge)
* Use dynamic versioning
* Bump `pydantic` to v2.11.4 in the lockfile, we have some tests failing
2025-05-15 17:39:14 -07:00
Nuno CamposandGitHub 909a4591a8 Port checkpointer/store fixtures in langgraph-prebuilt to idiomatic pattern (#4629) 2025-05-09 19:42:44 +00:00
Vadym BardaandGitHub 0e61cc2cf6 prebuilt: remove state_modifier (#4439)
Removes support for `state_modifier` param in `create_react_agent`. To
specify system prompt / instructions use `prompt` parameter.
2025-05-05 09:49:28 -04:00
Sydney RunkleandGitHub a0a302dec5 prebuilt: switch to executing parallel tool calls via Send by default (#4438) 2025-05-05 09:49:09 -04:00
Sydney Runkle 3cf2291354 docstrings for prebuilt 2025-05-01 13:40:58 -04:00
Vadym BardaandGitHub e25dde1df0 docs(reference): filter class methods and add missing docstrings (#4463) 2025-04-29 21:31:57 +00:00
Vadym BardaandGitHub ea55c2d468 docs: simplify docstring / API reference for create_react_agent (#4457) 2025-04-29 18:52:43 +00:00
Sydney Runkle 6250b364f7 version bumps and locks 2025-04-28 17:39:20 -04:00
vbarda 4b7ec256e9 prebuilt: switch to executing parallel tool calls via Send by default 2025-04-28 11:07:27 -04:00
Sydney Runkle 5e9bea5134 no upper bounds for python 2025-04-25 17:45:54 -07:00
Nuno Campos 70afd2dc2c Update all 2025-04-25 15:11:32 -07:00
Sydney Runkle 9e89b68596 fixing prebuilt tests 2025-04-22 17:24:54 -07:00
Vadym BardaandGitHub e89633f30b prebuilt: release 0.1.8 (#4161) 2025-04-03 12:01:18 -04:00
Vadym BardaandGitHub 67e8f8fc11 prebuilt: add optional pre-model hook that runs before calling LLM in create_react_agent (#4059)
Example:

```python
from typing import Any
from langchain_openai import ChatOpenAI
from langchain_community.tools.tavily_search import TavilySearchResults
from langchain_core.messages import AnyMessage
from langchain_core.messages.utils import count_tokens_approximately

from langgraph.graph import MessagesState
from langgraph.prebuilt.chat_agent_executor import create_react_agent, AgentState
from langgraph.checkpoint.memory import InMemorySaver

from langmem.short_term import SummarizationNode, RunningSummary


class State(MessagesState):
    context: dict[str, Any]

search = TavilySearchResults(max_results=3)
tools = [search]

model = ChatOpenAI(model="gpt-4o")
summarization_model = model.bind(max_tokens=256)

summarization_node = SummarizationNode(
    token_counter=count_tokens_approximately,
    model=summarization_model,
    max_tokens=2048,
    max_summary_tokens=256,
    output_messages_key="messages"
    # output_messages_key="llm_input_messages"
)

checkpointer = InMemorySaver()


class State(AgentState):
    user_language: str

    # summarization-related keys
    context: dict[str, Any]


def prompt(state):
    language = state["user_language"]
    system_msg = f"Always respond in {language}"
    return [{"role": "system", "content": system_msg}] + state["messages"]


graph = create_react_agent(
    model,
    tools,
    prompt=prompt,
    pre_model_hook=summarization_node,
    state_schema=State,
    checkpointer=checkpointer
)
```
2025-04-02 15:41:31 +00:00
Nuno Campos 49bb08a3f9 Update prebuilt test 2025-03-27 17:58:29 -07:00
Vadym BardaandGitHub e981d27f84 prebuilt: release 0.1.7 (#4034) 2025-03-26 09:22:09 -04:00
Vadym BardaandGitHub 71db4f2ad5 prebuilt: ignore updates when combining parent commands with Send (#4033) 2025-03-26 09:20:37 -04:00
Vadym BardaandGitHub f36b7f61fb prebuilt: release 0.1.6 (#4020) 2025-03-25 15:49:02 -04:00
Vadym BardaandGitHub 4095f0a927 prebuilt: only combine Command.PARENT for Send gotos in ToolNode (#4019) 2025-03-25 15:47:46 -04:00
Vadym BardaandGitHub af7515c37a prebuilt: release 0.1.5 (#4015) 2025-03-25 13:52:35 -04:00
Vadym BardaandGitHub b97cda4290 prebuilt: add support for multiple Command(graph=Command.PARENT) returned by tools (#4003) 2025-03-25 13:51:05 -04:00
Nuno Campos 2ee5b8e9d1 Bump 2025-03-24 09:03:23 -07:00
Nuno Campos 037d9d1402 Lock 2025-03-24 07:49:24 -07:00