cleanup-lead-owners
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from the HubSpot API and user-supplied employee lists, creating a surface for potential injection attacks.
- Ingestion points: Fetches CRM owner records via the requests library and ingests a list of current employees to identify records for reassignment.
- Boundary markers: No specific delimiters or safety instructions are defined to separate external data from agent instructions.
- Capability inventory: Possesses CRM write access via the Batch Update API and provides instructions for user deactivation, which are triggered by the ingested data.
- Sanitization: No mention of input validation or sanitization for owner names or employee list content.
- [EXTERNAL_DOWNLOADS]: The skill refers to an external tool for environment management.
- Evidence: Recommends the use of the uv tool from the astral-sh GitHub repository for managing the Python execution environment.
Audit Metadata