skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/claims-documentation/Gen Agent Trust Hub
claims-documentation
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The Python implementation in SKILL.md contains no malicious patterns, obfuscation, or unauthorized network operations. Its behavior matches its stated purpose of claim management.
- [PROMPT_INJECTION]: The skill processes user-provided narratives and descriptions which are reflected back to the agent, creating a risk of indirect prompt injection. Evidence: 1. Ingestion points: Project data and narratives provided via CSV, Excel, or JSON as specified in instructions.md. 2. Boundary markers: No delimiters or ignore instructions are used for user-provided text fields. 3. Capability inventory: The skill requires filesystem access and python3 for processing. 4. Sanitization: No input validation or filtering of external content is present in the implementation.
Audit Metadata