The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests local TypeScript source code and extracts JSDoc comments to include in the generated OpenAPI specification. While this involves processing untrusted data, the risk is minimal as the output is a structured JSON file for development use.
Ingestion points: scripts/generate-openapi.ts reads project source files using the TypeScript compiler API.
Boundary markers: None.
Capability inventory: fs.writeFileSync in scripts/generate-openapi.ts for saving the output spec.
Sanitization: JSDoc comments are extracted and inserted directly into the JSON description fields.
[SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration were detected. The skill's operations are confined to the local file system and are consistent with standard static analysis tools like linters or documentation generators.

bootgs-openapi