Back to Radar

LangChain

LangChain Intelligence

Name: LangChain bug intelligence
Creator: AI & Agent Dev Bug Sandbox

Published reports for this stack, with severity, affected version, and a direct path into each diagnostic.

Similar Reports

medium96%

pyproject.toml is read with locale encoding in unit tests, causing UnicodeDecodeError on non-UTF-8 Windows locales

Multiple unit test call sites read pyproject.toml without specifying encoding='utf-8'. On Windows with a non-UTF-8 locale (e.g., cp936), Python defaults to the ANSI code page, leading to UnicodeDecodeError because the file is valid UTF-8 and contains non-ASCII characters. Issue #40588 fixed two call sites, but two remain: one active in langchain_v1 summarization tests and one latent in langchain-classic dependencies tests.

Langchain

medium95%

langchain-classic: ImportError messages for moved REPL tools contain dead links and missing spaces

The ImportError messages raised when importing moved REPL tools/agents in langchain-classic contain outdated and dead links to SECURITY.md and a GitHub discussion, and have missing spaces causing link concatenation. This leads users to 404 pages or broken links, reducing safety guidance.

V1.0.8 Langchain-Classic

medium98%

Fix incorrect import paths in langchain-core docstring examples

Docstring examples in langchain-core reference modules 'langchain_core.output_parser', 'langchain_core.runnable', and 'langchain_core.documents' which do not exist, causing ModuleNotFoundError or ImportError when copied. The correct modules are 'langchain_core.output_parsers', 'langchain_core.runnables', and 'langchain_core.documents.base'.

V1.6.6 Langchain-Core

medium95%

langchain-classic: importing moved agent from agent_toolkits raises ValueError instead of ImportError

When importing a moved agent (e.g., create_csv_agent) from langchain_classic.agents.agent_toolkits, the __getattr__ helper incorrectly calls as_import_path without specifying relative_to, causing a ValueError from langchain_core instead of the intended ImportError with migration guidance.

V1.0.8 Langchain-Classic

medium75%

langchain-openai: named tool_choice rejected by Qwen thinking mode (DashScope)

ChatOpenAI with extra_body={"enable_thinking": true} and bind_tools(..., tool_choice="get_weather") sends a named tool-choice object to Alibaba DashScope, which rejects it with InternalError.Algo.InvalidParameter. LangChain should validate this incompatible combination before sending the request.

V1.6.6 LangChain

medium87%

@wrap_tool_call fails to preserve custom context type and is incompatible with create_agent(context_schema=...)

The @wrap_tool_call decorator does not accept or propagate a context_schema parameter, so the returned AgentMiddleware always types ContextT as None. When used with create_agent(context_schema=AgentContext), the invariant ContextT generic causes a type error, even though the runtime behavior may be correct.

V1.4.3 Langchain

medium85%

Feature Request: Pass `_meta` parameter through to MCP SDK `call_tool()`

langchain-mcp-adapters does not pass the MCP `_meta` field from tool arguments to the MCP SDK's `call_tool()`, preventing per-request metadata like session IDs, routing tags, or tenant context from reaching MCP servers.

Langchain

medium95%

core: `InMemoryRecordManager.list_keys(limit=0)` returns all keys instead of an empty list

The `limit` parameter is treated as falsy when `0`, causing all keys to be returned instead of an empty list for both `list_keys` and `alist_keys`.

V<=1.6.3 Langchain-Core

medium75%

Feature request: Add runtime security middleware documentation for agent tool interception (Deadend callback)

The issue requests adding documentation for integrating DeadendCallbackHandler, a third-party runtime circuit breaker, into LangChain agents to block prompt injection and RCE payloads at the on_tool_start phase before tool execution.

LangChain

medium95%

Add SiaMesh x402 Secure Scraper Tool

Feature request to integrate SiaMesh as a LangChain tool for secure web scraping with x402 micro-payments. The proposed solution adds a new SiaMeshTool class that wraps the public endpoint, enabling agents to scrape URLs while handling payment and anti-prompt-injection sanitization.

Langchain

medium65%

Proposed SiaMesh x402 tool introduces external service dependency and security risks without local fallback

The feature request proposes adding a SiaMeshTool to LangChain for secure scraping via the x402 payment protocol. While it addresses prompt injection and bot protection, the integration as described relies on a centralized external service with no local fallback, lacks input validation, and couples the tool to Solana payments.

Langchain

medium85%

ChatFireworks does not surface LangSmith Gateway metadata on successful responses

Fireworks integration lacks parsing of the X-LangSmith-Gateway-Metadata header on successful responses when using LANGSMITH_GATEWAY, causing missing gateway info and incorrect model/provider in traces.

Langchain

medium92%

Surface LangSmith gateway metadata on successful responses in ChatFireworks

ChatFireworks does not parse the X-LangSmith-Gateway-Metadata header on successful responses when routed through the LangSmith LLM Gateway, unlike ChatOpenAI and ChatAnthropic. This causes traces to miss gateway info such as the actually resolved model and provider, leading to incorrect cost tracking and inconsistent behavior across integrations.

LangChain

medium85%

Surface LangSmith gateway metadata on successful Fireworks completions

ChatFireworks does not capture the X-LangSmith-Gateway-Metadata header from successful gateway-routed responses, causing missing or incorrect model/provider information in LangSmith traces. This fix adds gateway metadata capture conditioned on the lsv2_ API key prefix, mirroring OpenAI and Anthropic integrations.

Langchain-Fireworks

low95%

Fix incorrect attribution of macOS Seatbelt to Anthropic in CodexSandboxExecutionPolicy docstring

Documentation-only bug in LangChain: the docstring for CodexSandboxExecutionPolicy incorrectly states that Seatbelt (macOS) is provided by Anthropic. In reality, Seatbelt is Apple's macOS sandbox facility and Codex CLI is OpenAI's tool. The proposed fix updates the phrase to accurately reflect the technology and vendor.

V1.4.0 Langchain