lead-research-assistant
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it ingests and processes untrusted data from the open web and external repositories to identify and qualify leads.
- Ingestion points: The skill analyzes external content including company websites, job postings, recent news, and public GitHub repositories (specifically mentioned in the examples) to find signals of company needs and technology stacks.
- Boundary markers: The instructions do not define any delimiters or system-level warnings to ignore instructions that might be embedded within the researched external data.
- Capability inventory: The skill primarily generates informational reports and outreach strategies. While it suggests saving results to a CSV for CRM import, it does not currently exhibit direct shell execution or arbitrary file-write capabilities.
- Sanitization: There is no evidence of sanitization, filtering, or escaping of the external content before it is processed by the agent or presented to the user.
Audit Metadata