The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted code from the repository diff, which could potentially contain malicious instructions or adversarial content designed to influence the agent's output.\n- Ingestion points: The skill uses git diff --cached to read the content of staged changes (SKILL.md).\n- Boundary markers: No explicit markers or instructions are provided to delineate the diff content or to instruct the agent to ignore any embedded instructions within that content.\n- Capability inventory: The skill has the capability to generate shell commands (git commit) for the user to execute.\n- Sanitization: There is no evidence of sanitization or escaping mechanisms applied to the diff data before it is processed by the agent.

git-commit-message