The Agent Skills Directory

[SAFE]: The skill implements a standard software engineering workflow for task decomposition. It defines clear templates for 'vertical slices' and 'wide refactors' using expand-contract patterns. The use of local file paths (e.g., .scratch/) and external tracker integration is driven by user-approved instructions and explicit configuration (referencing a setup skill). No malicious patterns such as obfuscation, credential harvesting, or unauthorized remote code execution were detected. The skill uses 'disable-model-invocation: true' in its frontmatter, which is a restrictive configuration, and follows a human-in-the-loop process (Step 4: Quiz the user) before taking any action.

to-tickets