The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data (diffs and guidelines) that could contain malicious instructions designed to influence the agent's behavior during the review process.
Ingestion points: The agent reads the content of project guideline files (e.g., docs/TEST_GUIDELINES.md) and the output of git diff commands.
Boundary markers: The instructions do not specify any delimiters or safety headers to separate the instructions of the skill from the content of the data being analyzed.
Capability inventory: The skill uses git diff via subprocess execution to retrieve and process changed lines of code.
Sanitization: There is no evidence of sanitization, filtering, or validation of the input data before it is processed by the agent.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to retrieve information from the repository.
Evidence: Step 1 requires the execution of git diff <range> -- <path>... to analyze changes in the codebase.

review-conformance