The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes dynamic context injection in SKILL.md to execute shell commands such as git status, git diff, git branch, and git log at load time. These operations are used to populate the agent's context with relevant repository information and are benign in this workflow.
[COMMAND_EXECUTION]: To ensure commit quality, the skill instructs the agent to run project-specific scripts including pnpm build, pnpm test, and various linting tools (pnpm lint:eslint, etc.). This is the primary intended behavior of the skill and follows standard development practices.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from the repository's git history and file diffs.
Ingestion points: Real-time git status, diff, and log output captured in SKILL.md via dynamic context injection.
Boundary markers: Absent; the shell output is interpolated directly into the agent's context without delimiters or instructions to ignore embedded content.
Capability inventory: The skill has the capability to stage files and execute various shell commands through pnpm and git as described in SKILL.md.
Sanitization: No sanitization or escaping of the ingested git data is performed prior to processing.

commit