setup-company

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill facilitates the collection of sensitive business and personal data, including pricing and professional backgrounds, into a single local Markdown file (company-profile.md). While this centralizes sensitive data, no automated exfiltration logic or external network transmission of this data was detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill utilizes external web research to pre-fill profile data, which presents an ingestion surface for potential indirect instructions. 1. Ingestion points: User-provided URLs for company websites, documentation, and knowledge bases. 2. Boundary markers: None are specified to isolate fetched content. 3. Capability inventory: The agent is instructed to perform web research/browsing. 4. Sanitization: No automated sanitization is described, although the agent is required to confirm drafted content with the user before saving the file.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 01:34 PM