accelint-architecture-doc
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository files to generate documentation, creating a surface for embedded instructions to potentially influence agent outputs.\n
- Ingestion points: Stage 2 scans project artifacts including
README.md,package.json,docker-compose.yml, infrastructure configs, and CI workflows.\n - Boundary markers: The skill uses a structured 11-section markdown template to organize information, which provides implicit structural boundaries.\n
- Capability inventory: The agent has permissions to read codebase files and write to
ARCHITECTURE.md,AGENTS.md, andCLAUDE.md.\n - Sanitization: The workflow requires a manual preview and user confirmation checkpoint before any filesystem modifications occur.\n- [EXTERNAL_DOWNLOADS]: The README provides installation instructions that reference the vendor's official GitHub repository.\n
- Evidence:
https://github.com/gohypergiant/agent-skills
Audit Metadata