competitor-matrix

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill uses WebSearch and WebFetch to ingest data from external competitor websites and pricing pages. This data is used to generate a matrix that is saved to the local file system (founder/competitor-matrix.md). There is a risk that a competitor's website could contain hidden instructions designed to manipulate the agent's output or perform unauthorized file operations.
  • Ingestion points: External competitor homepages, pricing pages, and funding news sources accessed via WebSearch and WebFetch in the 'Market landscape' and 'Feature comparison matrix' steps.
  • Boundary markers: Absent. The instructions do not define delimiters for the external content or provide explicit warnings to the agent to disregard instructions found within the fetched data.
  • Capability inventory: The skill has Edit(founder/**) permissions to write reports to the disk and Read permissions to access sensitive startup data in founder/facts.md and founder/product-brief.md.
  • Sanitization: Absent. The skill instructions do not specify any validation or filtering of the content retrieved from external sources before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 02:49 AM