The Agent Skills Directory

[SAFE]: The skill implements a standard developer workflow for git version control. It promotes security by requiring a manual review of staged changes and specifically checking for the presence of hardcoded secrets, tokens, or accidental debug logs.
[INDIRECT_PROMPT_INJECTION]: The skill reads project files via git commands to identify changes, which creates a theoretical attack surface where malicious instructions could be embedded in the code being committed. This is an inherent functional requirement for the skill's operation.
Ingestion points: git status and git diff commands in SKILL.md and README.md.
Boundary markers: Absent; the agent processes raw git output.
Capability inventory: git commit and project-specific verification commands (lint, test, build) as described in SKILL.md.
Sanitization: None; the skill is designed to accurately commit the user's local modifications.

commit-work