The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local git commands to manage repository state.
Evidence: The workflow in SKILL.md uses git diff --cached --stat, git diff --cached, git log --oneline -5, and git commit -m "<message>" to perform its primary function.
[INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data from the repository, which could contain malicious instructions embedded in file content or comments.
Ingestion points: Data enters the agent's context through git diff --cached --stat, git diff --cached, and git log as described in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or specific "ignore" markers for the content read from the diffs.
Capability inventory: The skill has the capability to execute git commit to modify the repository history.
Sanitization: The skill implements a security check in SKILL.md (Step 2) to specifically look for secrets, credentials, tokens, or .env files and abort the process if they are found.

commit