competitor-displacement-targeting

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it fetches content from external, untrusted websites to identify target companies and details.\n
  • Ingestion points: The skill uses WebFetch in step 3 to retrieve content from third-party sites like G2, Capterra, and competitor case study pages.\n
  • Boundary markers: No explicit delimiters or instructions are used to separate retrieved web content from the agent's core instructions, nor are there commands to ignore instructions embedded within the fetched data.\n
  • Capability inventory: Extracted data is used to drive subsequent automated actions, including company enrichment (enrich_company), people searching (search_people), and the creation of persistent lead lists (create_lead_list).\n
  • Sanitization: There is no evidence of sanitization or validation of the web content before it is processed for entity extraction.\n- [EXTERNAL_DOWNLOADS]: The skill performs automated searches and fetches content from various third-party domains (e.g., g2.com, capterra.com, stackshare.io, builtwith.com) to identify and document competitor usage patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 11:36 PM
Security Audit — agent-trust-hub — competitor-displacement-targeting