anysite-lead-generation

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, creating a potential surface for indirect prompt injection attacks.\n
  • Ingestion points: Data is ingested through the webparser tool (which parses arbitrary URLs) and LinkedIn search/enrichment tools via mcp__anysite__execute (e.g., source: "linkedin" or source: "webparser").\n
  • Boundary markers: The instructions do not define clear delimiters or provide instructions to the agent to ignore potential commands embedded in the scraped content, increasing the risk of the agent following instructions found within prospect data.\n
  • Capability inventory: The skill possesses the ability to export data via mcp__anysite__export_data (which generates download URLs for CSV/JSON exports) and displays fetched data directly in the chat context.\n
  • Sanitization: There is no evidence of sanitization, filtering, or validation of the external data before it is presented to the agent's context or exported to files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 12:14 PM