inbound-lead-qualification

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from external sources that may contain malicious instructions intended to manipulate the agent's behavior.
  • Ingestion points: The skill parses raw lead data from various formats including CSVs and pasted text (Step 1), and it specifically analyzes user-generated intent data such as demo request messages and chatbot conversations (Step 5).
  • Boundary markers: Absent; there are no instructions or delimiters provided to the agent to treat external lead data as untrusted or to ignore embedded instructions within that data.
  • Capability inventory: The agent has access to CRM systems (HubSpot, Salesforce), local file system read/write permissions for configuration and CSV generation, and web search capabilities for company research.
  • Sanitization: Absent; the skill does not describe any validation, filtering, or escaping of the content extracted from leads before it is used for scoring and qualification logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 10:48 AM