inbound-lead-qualification
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted external data in the form of inbound lead details, including free-text fields like 'demo request messages', 'chatbot conversations', and 'downloaded content' titles.
- Ingestion points: Untrusted data enters the context via the
inbound-lead-datarequirement and the manual parsing of lead lists in Step 1. - Boundary markers: The instructions do not define clear delimiters or use 'ignore embedded instructions' warnings when processing the inferred intent from lead messages.
- Capability inventory: The skill possesses
web-search,crm-lookup, andfile-write(CSV output) capabilities, which could be influenced by malicious lead data. - Sanitization: There are no explicit instructions for sanitizing or escaping external content before the agent interpolates it into the qualification reasoning steps.
Audit Metadata