factory-files
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill reads local configuration files and transmits their contents to the vendor's validation API at app.warp.dev for checking. It also forwards the WARP_API_KEY environment variable to the remote server for authentication purposes when present.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local repository files, presenting a potential attack surface.
- Ingestion points: The scripts/validate_factory_files.py script reads file content from various resource files within the local directory tree.
- Boundary markers: Instructions specify the use of Markdown fences and frontmatter blocks to isolate data and prompts.
- Capability inventory: The skill can read local files and perform network requests to external vendor APIs.
- Sanitization: The validation script includes checks to prevent following symlinks that resolve outside the factory root to protect against path traversal attacks.
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill fetches JSON schemas from a remote vendor URL (app.warp.dev) at runtime using curl to ensure validation rules are up to date.
Audit Metadata