The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests content from the local repository which may contain untrusted data that could influence the agent's behavior.
Ingestion points: The skill uses git diff --cached and git diff --cached --stat in SKILL.md to read staged code changes.
Capability inventory: The skill has the ability to execute git commit based on its analysis.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the code diff as untrusted data or to ignore instructions embedded within the code.
Sanitization: The skill does not perform sanitization or filtering on the diff content before processing it.
[COMMAND_EXECUTION]: The skill executes standard system commands to interact with the version control system.
Evidence: Use of git diff, git status, and git commit in SKILL.md to manage repository state.

formatting-commit-messages