skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/material-tracker/Gen Agent Trust Hub
material-tracker
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data in formats like CSV, Excel, and JSON, which introduces a surface for indirect prompt injection attacks where malicious instructions could be embedded in the project data.
- Ingestion points: The instructions.md file specifies that the agent should gather and process project data from common external formats (CSV, Excel, JSON).
- Boundary markers: No explicit boundary markers or delimiters are defined in the instructions to separate user-provided data from agent instructions.
- Capability inventory: The skill includes the capability to write to the file system using the export_to_excel method in SKILL.md, and requires the filesystem permission as noted in claw.json.
- Sanitization: There is no evidence of input sanitization or validation logic within the Python implementation in SKILL.md to filter out potentially malicious content from the processed data.
Audit Metadata