The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes git diff and git log to analyze repository changes. It passes a user-supplied 'fixed point' (such as a branch name or commit SHA) directly into these shell commands.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the codebase being reviewed.
Ingestion points: SKILL.md (Step 1 and Step 2) retrieves data from git diff, git log, and external issue trackers specified in commit messages.
Boundary markers: The instructions do not define clear delimiters or use 'ignore embedded instructions' warnings when passing diff content to sub-agents.
Capability inventory: The skill can execute shell commands (git) and has the ability to read various configuration, documentation, and source files across the repository.
Sanitization: There is no evidence of sanitization or filtering for malicious instructions that might be embedded in code comments or commit messages before they are processed by the sub-agents.

review