The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a bundled Node.js script scripts/diff-baseline.mjs that spawns git subprocesses to analyze tracked and untracked files in the repository for change verification.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external project data to generate walkthroughs, creating an attack surface for indirect instructions. 1. Ingestion points: Acceptance criteria, latest requests, and file diffs accessed as specified in SKILL.md. 2. Boundary markers: None; there are no instructions or delimiters to prevent the agent from following instructions embedded in the analyzed artifacts. 3. Capability inventory: The skill possesses file system read/write capabilities and git command execution through its baseline scripts. 4. Sanitization: Instructions recommend redacting sensitive data like PII from screenshots but do not provide sanitization for instruction-based content in text inputs.

walkthrough