Last commit on October 2, 2026·Created on January 22, 2026
langchain-ai/langchain-skills
“A framework for constructing Deep Agents with structured harnesses and persistent state”
Combined rank
#304
across all skills
Stars
1.3k
+1.3% in last 7d
Forks
105
+4.0% in last 7d
Watchers
9
0.0% in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
TL;DR
This collection provides the architectural patterns necessary for building Deep Agents applications. It focuses on the implementation of agent harnesses and the SKILL.md format to ensure consistent agent behavior.
Developers can implement sophisticated memory systems using ephemeral state backends and persistent store backends, enabling agents to maintain context across sessions or interact with the filesystem via specialized middleware.
WHO IT'S FOR
AI agent / automation builders
building Deep Agents applications with custom harnesses
Backend / infrastructure engineers
implementing agent memory and persistent state backends
Technical leads / architects
designing multi-agent orchestration and HITL workflows
Python developers
scaffolding minimal local Deep Agents in Python
Claude
Codex
Cursor
Gemini CLI
Copilot
WindsurfWant to install it?
Claude Marketplace · Claude Plugin · Clone to Codex · Manual Copy · NPX Skills AddRepository instructions