The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the git and gh (GitHub CLI) command-line interfaces to inspect the repository status and manipulate commit history. These operations are essential to the primary function of performing fixup commits and are consistent with professional development workflows.\n- [PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection by processing external data from the local repository environment.\n
Ingestion points: The skill reads commit history through git log, commit content via git show, and local guidelines from .claude/rules/commit-message.md.\n
Boundary markers: The instructions do not specify delimiters to isolate the ingested text from the agent's core instructions.\n
Capability inventory: The skill possesses the ability to modify the local file system (via git commits) and trigger other functional skills like /commit or /squash.\n
Sanitization: There is no evidence of sanitization or filtering applied to the commit data or rule files before they are processed by the agent. This represents a standard operational risk for tools that analyze developer-controlled content.

fixup