market-landscape-scan

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process information from external OSINT sources which could potentially include malicious instructions embedded in web content.
  • Ingestion points: External research sources including practitioner forums, review sites, and industry press mentioned in the workflow (SKILL.md).
  • Boundary markers: The skill implements a search-plan gate and requires a mandatory 'Fact/Inference/Assumption' labeling system to distinguish between raw data and agent reasoning.
  • Capability inventory: The skill relies on standard search and browsing tools to collect intelligence across multiple disciplines (OSINT, FININT).
  • Sanitization: Includes a 'Do-not-invent list' to prevent the fabrication of sensitive market data and requires direct citations for all factual claims.
  • [EXTERNAL_DOWNLOADS]: The skill includes a reference to an external GitHub repository for methodological inspiration.
  • Source: https://github.com/deanpeters/product-manager-prompts (SKILL.md).
  • Context: The link is provided as a reference for adapted prompt engineering techniques and does not involve automated downloads or execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 02:56 PM
Security Audit — agent-trust-hub — market-landscape-scan