The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from git diff and git log outputs. Malicious instructions embedded within code changes or commit history could potentially influence the agent's behavior during the message generation process.
Ingestion points: SKILL.md and SKILL-CN.md workflow steps (e.g., git diff, git log, and reading convention files like CONTRIBUTING.md).
Boundary markers: Absent; the instructions do not provide the agent with explicit delimiters or instructions to treat the diff/log content as non-instructional data.
Capability inventory: The skill is restricted to local repository inspection commands (git diff, git log, git status). No network or administrative write capabilities are requested in the instructions.
Sanitization: There is no evidence of sanitization or filtering logic to prevent the execution of instructions contained within the analyzed repository data.
[COMMAND_EXECUTION]: The skill instructs the agent to execute standard git commands to inspect the repository state.
Evidence: Use of git diff, git log, and git status in the SKILL.md workflow to gather context for commit message generation.

git-commit