csv-analyse
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from CSV files and processes it to generate summaries. While this represents an attack surface for indirect prompt injection where malicious instructions could be embedded in cell values, the script outputs structured JSON, which helps the agent distinguish between data and instructions.
- Ingestion points: scripts/analyse.py reads CSV content from sys.stdin.
- Boundary markers: None explicitly mentioned in the skill instructions to separate data from instructions.
- Capability inventory: The skill employs fs_list, fs_read, and run_skill_script.
- Sanitization: Data is parsed via pandas.read_csv; the agent receives structured summaries rather than raw text.
- [EXTERNAL_DOWNLOADS]: The skill documentation identifies a dependency on the pandas library, which is expected to be available in the Cloudflare Sandbox environment.
Audit Metadata