crm-cleanup
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from a CRM system that could potentially contain malicious payloads designed to influence agent behavior.
- Ingestion points: Data is retrieved from the
~~crmconnector, including deal details, contact information, and notes (SKILL.md). - Boundary markers: There are no explicit delimiters or specific instructions for the agent to ignore instructions embedded within the CRM data.
- Capability inventory: The skill is granted access to the
Bash,WebFetch, andReadtools, which could be abused if an injection is successful (frontmatter). - Sanitization: The skill mitigates risks through mandatory "approval gates," requiring explicit user confirmation for every modification and using side-by-side diffs for review (SKILL.md).
Audit Metadata