The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run 'git diff main...HEAD --name-only'. This is a standard read-only command used to identify modified files for review and is considered safe within this context.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted content from the repository, including specification files and modified source code. Ingestion points: Files in 'context/specs/', 'context/architecture.md', 'context/code-standards.md', 'context/ui-registry.md', and modified files. Boundary markers: None identified. Capability inventory: Reading local files and generating reports. Sanitization: No sanitization is performed on ingested content.

code-review