The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git (to diff local work) and the GitHub CLI (gh) to retrieve PR information. These are standard operations for a development-focused tool.\n- [EXTERNAL_DOWNLOADS]: Fetches Pull Request data and diffs from GitHub. These downloads originate from a well-known service and are required for the skill's primary purpose.\n- [PROMPT_INJECTION]: The skill processes external data (such as COMMANDMENTS.md files and PR descriptions) to use as guidelines for auditing. This creates a theoretical surface for indirect prompt injection, but it is intrinsic to the skill's role as a conventions auditor.\n
Ingestion points: Loads text from COMMANDMENTS.md (root, docs, or home directory) and PR metadata from GitHub.\n
Boundary markers: The skill instructions do not specify explicit delimiters or "ignore previous instructions" guards for the loaded text.\n
Capability inventory: The agent has access to Bash, Read, Grep, Glob, Edit, and Agent tools to audit code and apply optional fixes.\n
Sanitization: No specific filtering or validation of the guideline content is described.

commandments