inbound-lead-triage
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources that could contain malicious instructions designed to manipulate the agent's logic or behavior.
- Ingestion points: Raw lead data is ingested from CRM forms, CSV exports, and chatbot conversations, specifically within fields like
raw_messageandconversation_summaryin theinbound-lead-datarequirement. - Boundary markers: The skill does not define specific delimiters or instructions for the agent to ignore potentially adversarial commands embedded within the lead content.
- Capability inventory: The agent has access to web search, contact enrichment tools, file system writes (saving CSVs and config files), and CRM integrations. Maliciously crafted lead data could attempt to trigger these tools in unintended ways (e.g., forcing a high-priority tier for a specific email or triggering specific web searches).
- Sanitization: There is no evidence of sanitization, filtering, or escaping for the external content before it is processed by the triage and enrichment logic.
Audit Metadata