The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes git diff and git log to retrieve the codebase changes and history between a user-defined point and the current HEAD. These commands are essential for the primary functionality of reviewing code changes.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the repository being reviewed.
Ingestion points: The skill reads output from git diff and git log, as well as the contents of specification files found in directories like docs/, specs/, or .scratch/ and standard files like CLAUDE.md.
Capability inventory: The extracted data is passed to parallel general-purpose sub-agents for logical analysis and reporting.
Boundary markers: Absent. The prompts provided to the sub-agents do not employ specific delimiters or instructions to ignore or isolate potential malicious prompts embedded within the code or specifications.
Sanitization: No sanitization or filtering of the external content is performed before it is interpolated into the prompts for the sub-agents.

review