tjdwls101010/skills-for-langchain (langchain)

Consultant and current-API guide for building agents with the LangChain ecosystem (LangChain 1.x, LangGraph 1.x, Deep Agents / `deepagents`), Python. Load this whenever a user wants to design or build an agent, automate a multi-step task, build an assistant or chatbot, or have something answer from their data — even when they describe only an abstract goal, name no framework, and show no code — and lead them from that goal to a concrete architecture and, on agreement, an implementation. Also load whenever writing, reviewing, or editing Python that imports LangChain, LangGraph, or Deep Agents; calls `create_agent` or `create_deep_agent`; or works with agent middleware, subagents, backends, long-term memory, human-in-the-loop, permissions, streaming, structured output, or agent deployment — to supply current APIs the model otherwise gets wrong, especially for Deep Agents. This skill carries recent deltas such as `create_deep_agent(system_prompt=...)` instead of `instructions=`, `create_agent` instead of `create_react_agent`, and the current agent-as-tool pattern. Do not use for CrewAI, AutoGen, LlamaIndex, or raw OpenAI or Anthropic SDK work unless it is explicitly bridged through LangChain.

Claude Skills are markdown + scripts loaded into Claude Code; scanned with the same pipeline.

Source:

github.com/tjdwls101010/skills-for-langchain

Provenance:

Found by crawling (provenance, not a security rating)

We found this by crawling public sources and nobody has attested to it. That is how most of the registry gets here, and it is not a mark against the artifact — it only means no authenticated act of publication is on record. An artifact can be the vendor's own and still appear here.

70 critical findings reported by cisco-skill-scanner. The composite score alone does not reflect this — read the per-scanner reports below.