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.
This skill enables AI agents to capture learnings, failed commands, and user corrections in markdown files. By documenting missing capabilities or outdated knowledge during a session, the agent creates a structured record of improvements to reference before starting future tasks.
WHO IT'S FOR
AI agent / automation builders
implementing continuous learning loops for agents
OpenClaw users
enhancing agent memory and knowledge management
Technical leads / architects
standardizing AI agent error handling and feedback
AI agent power users
reducing repetitive AI mistakes in long-term projects
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
The project or workspace root where the learning should be saved.
A correction describing the mistaken assumption and the correct information.
Use the self-improving-agent skill to record the correction I provide in this project's .learnings/LEARNINGS.md with category correction. First ensure the learning directory and log files exist without overwriting existing files. Keep the entry concise and exclude secrets.
Claude
Codex
OpenClaw
Copilot
Amp