gtm-competitor-gap
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from web search results and Notion content. While this creates a potential surface for indirect prompt injection, it is assessed as safe because the skill lacks the capabilities (such as command execution or arbitrary file access) to exploit the ingested data. \n
- Ingestion points: Reads market/ICP and sizing data from Notion; retrieves competitor details, prices, and reviews from external sources via
WebSearch(SKILL.md). \n - Boundary markers: None explicitly defined in the instructions to separate external content from agent instructions. \n
- Capability inventory: No subprocess execution, dynamic code execution (eval/exec), or sensitive file system access tools are utilized. \n
- Sanitization: No specific data sanitization or filtering steps are defined for the external content. \n- [SAFE]: The skill adheres to its stated purpose of competitor analysis without attempting credential theft, remote code execution, persistence, or privilege escalation. All external tool usage is for legitimate research purposes.
Audit Metadata