The agent engineering platform.
Python agentsaiai-agentsanthropicchatgptdeepagentsenterpriseframeworkgeminigenerative-ailangchainlanggraphllmmultiagentopen-sourceopenaipydanticpythonragtypescript
docs(anthropic): cleanup nits (#34299)
Mason Daughertyae1f03fbe04d
diff --git a/libs/partners/anthropic/langchain_anthropic/chat_models.py b/libs/partners/anthropic/langchain_anthropic/chat_models.pyindex 2857bb9141..d85559e531 100644--- a/libs/partners/anthropic/langchain_anthropic/chat_models.py+++ b/libs/partners/anthropic/langchain_anthropic/chat_models.py@@ -94,7 +94,12 @@ _FALLBACK_MAX_OUTPUT_TOKENS: Final[int] = 4096class AnthropicTool(TypedDict):- """Anthropic tool definition."""+ """Anthropic tool definition for custom/user-defined tools.++ Custom tools use `name` and `input_schema` fields to define the tool's+ interface. These are converted from LangChain tool formats (functions, Pydantic+ models, BaseTools) via `convert_to_anthropic_tool`.+ """name: str@@ -113,8 +118,25 @@ class AnthropicTool(TypedDict):allowed_callers: NotRequired[list[str]]-# Some tool types require specific beta headers to be enabled-# Mapping of tool type patterns to required beta headers+# ---------------------------------------------------------------------------+# Built-in Tool Support+# ---------------------------------------------------------------------------+# When Anthropic releases new built-in tools, two places may need updating:+#+# 1. _TOOL_TYPE_TO_BETA (below) - Add mapping if the tool requires a beta header.+# Not all tools need this; only add if API returns a beta header error.+#+# 2. _is_builtin_tool() - Add the tool type prefix to _builtin_tool_prefixes.+# This ensures the tool dict is passed through to the API unchanged (instead+# of being converted via convert_to_anthropic_tool, which may fail).+#+# 3. langchain_anthropic/tools/ - Add a factory function for better devx+# (IDE autocomplete, type checking, documentation).+# NOT required for the tool to work - users can always pass raw dicts.+# ---------------------------------------------------------------------------++# Some tool types require specific beta headers to be enabled.+# Mapping of tool type to required beta header._TOOL_TYPE_TO_BETA: dict[str, str] = {"web_fetch_20250910": "web-fetch-2025-09-10","code_execution_20250522": "code-execution-2025-05-22",@@ -138,7 +160,7 @@ _ANTHROPIC_EXTRA_FIELDS: set[str] = {def _is_builtin_tool(tool: Any) -> bool:"""Check if a tool is a built-in Anthropic tool.- [Claude docs for built-in tools](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview)+ [Claude docs](https://platform.claude.com/docs/en/agents-and-tools/tool-use/overview)"""if not isinstance(tool, dict):return False@@ -1134,7 +1156,7 @@ class ChatAnthropic(BaseChatModel):!!! example- ```python hl_lines="5-6"+ ```python hl_lines="6"from langchain_anthropic import ChatAnthropicmodel = ChatAnthropic(@@ -1366,6 +1388,7 @@ class ChatAnthropic(BaseChatModel):],}]+response = model.invoke(messages)response.content```@@ -1464,8 +1487,7 @@ class ChatAnthropic(BaseChatModel):See [`ChatAnthropic.with_structured_output()`][langchain_anthropic.chat_models.ChatAnthropic.with_structured_output]for more info, including strict output validation.- ```python hl_lines="13"- from typing import Optional+ ```python hl_lines="12"from pydantic import BaseModel, Field@@ -1855,7 +1877,7 @@ class ChatAnthropic(BaseChatModel):"""List of beta features to enable. If specified, invocations will be routedthrough `client.beta.messages.create`.- Example: `#!python betas=["mcp-client-2025-04-04"]`+ Example: `#!python betas=["token-efficient-tools-2025-02-19"]`"""# Can also be passed in w/ model_kwargs, but having it as a param makes better devx#@@ -2957,7 +2979,7 @@ class ChatAnthropic(BaseChatModel):method = "json_schema"if method == "function_calling":- formatted_tool = convert_to_anthropic_tool(schema)+ formatted_tool = cast(AnthropicTool, convert_to_anthropic_tool(schema))tool_name = formatted_tool["name"]if self.thinking is not None and self.thinking.get("type") == "enabled":llm = self._get_llm_for_structured_output_when_thinking_is_enabled(@@ -3121,7 +3143,7 @@ def convert_to_anthropic_tool(Requires Claude Sonnet 4.5 or Opus 4.1.Returns:- An Anthropic tool definition dict.+ `AnthropicTool` for custom/user-defined tools"""# already in Anthropic tool formatif isinstance(tool, dict) and all(diff --git a/libs/partners/anthropic/uv.lock b/libs/partners/anthropic/uv.lockindex b4274b1ae5..01f15653a4 100644--- a/libs/partners/anthropic/uv.lock+++ b/libs/partners/anthropic/uv.lock@@ -643,7 +643,7 @@ typing = [[[package]]name = "langchain-core"-version = "1.1.2"+version = "1.1.3"source = { editable = "../../core" }dependencies = [{ name = "jsonpatch" },