The Agent Skills Directory

Command Execution: The instructions in references/review-checklist.md and references/routing-analysis.md suggest running npm run tokens and npm test. These commands are standard development practices for validating code and measuring resource usage within a local repository.
Indirect Prompt Injection Surface: This skill is designed to ingest and analyze external code and documentation from Pull Requests, which introduces a potential surface for indirect prompt injection. 1. Ingestion points: The skill reads changed files in a PR (SKILL.md, references, tests, scripts) as primary data for its review workflow in SKILL.md. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded instructions are defined for the analyzed files. 3. Capability inventory: The skill utilizes shell commands (npm) for validation as documented in references/review-checklist.md. 4. Sanitization: No explicit sanitization or filtering of the input file content is specified before the agent performs its analysis.

skill-reviewer