The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute standard git commands to gather context for the commit message.
Evidence: git diff, git diff HEAD, and git log --oneline -40 are used to inspect the repository state.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data (source code changes) which could contain hidden instructions intended to manipulate the agent's output or behavior.
Ingestion points: git diff and git log output are read directly into the agent's context (SKILL.md).
Boundary markers: Absent. There are no delimiters or instructions provided to differentiate between code data and natural language instructions.
Capability inventory: The skill utilizes shell commands for reading repository data but explicitly states it is a "read-only operation."
Sanitization: Absent. The skill does not perform any validation or filtering of the content retrieved from the git commands.

aicommit