lead-generation

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates interaction with 'mcp.exa.ai', the official Model Context Protocol endpoint for Exa. This connection is used to fetch research data and structured lead information as part of its primary function.
  • [COMMAND_EXECUTION]: Uses Bash and Python's built-in 'csv' module to generate lead lists. This is a standard operation for data processing and export tasks within this workflow.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by interpolating user-provided inputs like company names and ICP descriptions into search queries. Evidence: found in Step 1 and Step 2 of SKILL.md. Capability inventory: 'Bash' and 'Write' tools are available for file output. Sanitization: The skill recommends using 'csv.writer' to handle data escaping. Boundary markers: No explicit delimiters are used for user input interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 02:08 AM
Security Audit — agent-trust-hub — lead-generation