skills/datadrivenconstruction/ddc_skills_for_ai_agents_in_construction/df-merger/Gen Agent Trust Hub
df-merger
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from external sources like CSV, Excel, and JSON files, which creates a surface for indirect prompt injection attacks where malicious instructions could be embedded in the data cells.
- Ingestion points: Data input is accepted through file paths or direct input as described in instructions.md and SKILL.md.
- Boundary markers: The skill lacks explicit markers or safety instructions to distinguish between data content and agent instructions.
- Capability inventory: The skill is restricted to pandas-based data manipulation and lacks high-risk capabilities such as network exfiltration or subprocess spawning.
- Sanitization: No sanitization is performed on the data content to detect or neutralize potential prompt injection patterns.
Audit Metadata