The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes external data from the repository to generate commit messages and update documentation.
Ingestion points: Repository diffs (git diff), plan files (PLAN.md, .plans/), and subdirectory documentation (README files).
Boundary markers: Absent; the instructions do not define delimiters or isolation markers for the ingested repository content.
Capability inventory: The skill can execute git add, git commit, and git reset --soft, and can perform file writes to documentation files.
Sanitization: Absent; no validation or escaping of the repository data is performed before it is processed by the agent.
[COMMAND_EXECUTION]: The skill executes local git commands to manage code changes. It includes safety guardrails that prohibit destructive operations like git reset --hard, git clean, and unapproved git push. However, the asap mode uses git commit -n to bypass pre-commit hooks, which may circumvent local repository security or quality validation workflows.

commit