outbound-prospecting-engine
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing untrusted data to generate outreach messaging.\n
- Ingestion points: Data is ingested from external sources such as job postings, LinkedIn posts, and conference attendee lists (Step 1 and Step 2 in SKILL.md).\n
- Boundary markers: The instructions lack explicit boundary markers or warnings to the agent regarding the untrusted nature of the signal data gathered from the web.\n
- Capability inventory: The system includes capabilities to send emails (agentmail) and launch marketing campaigns (setup-outreach-campaign), which could be abused if the agent follows instructions hidden in the ingested data.\n
- Sanitization: No sanitization or validation logic is specified for the data retrieved from external sources before it is used for personalization in Step 6.
Audit Metadata