The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to run various shell commands for auditing tasks, including dependency analysis with npx madge and circular dependency checks with deptrac.
[EXTERNAL_DOWNLOADS]: The skill utilizes npx, which downloads packages from the official npm registry, a well-known service, when running the madge tool.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes multiple project documentation files, which constitutes a potential attack surface for indirect prompt injection.
Ingestion points: Reads local files including PRD.md, spec.md, TASKS.md, SECURITY_CHECKLIST.md, and PROJECT_STATE.md.
Boundary markers: The instructions do not define specific delimiters to isolate the content of audited files from the agent's internal logic.
Capability inventory: The agent is granted capabilities for file system reads and shell command execution across all verification dimensions.
Sanitization: No explicit sanitization or validation of the ingested file content is performed before processing.

integrity-check