The Agent Skills Directory

[SAFE]: The skill operates entirely within the workspace to improve AI agent performance by documenting and fixing recurring errors.
[SAFE]: It does not perform network operations, download external code, or access sensitive credentials.
[SAFE]: The skill implements clear boundaries and delegates security-sensitive tasks (like editing other skills) to a specialized author-skill rather than attempting to modify executable code directly.
[SAFE]: All instructional content follows standard postmortem and continuous improvement practices (e.g., blameless root cause analysis).

continuous-learning