skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/project-closeout-checklist/Gen Agent Trust Hub
project-closeout-checklist
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes project data and file paths provided by the user in various formats (CSV, Excel, JSON), which introduces a surface for potential indirect prompt injection attacks.\n
- Ingestion points: The agent is instructed to accept and process data provided by the user or referenced in the skill as described in instructions.md.\n
- Boundary markers: The instructions lack specific delimiters or instructions for the agent to ignore embedded commands within the processed data.\n
- Capability inventory: The skill has filesystem access to export results to Excel using the pandas library, but it does not possess network access, shell execution, or privilege escalation capabilities.\n
- Sanitization: While the instructions specify to "validate inputs before processing," no concrete sanitization or filtering logic is implemented in the Python code in SKILL.md.
Audit Metadata