market-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to gather data from untrusted external sources such as industry blogs and review sites, which could contain malicious prompts designed to manipulate the agent.
- Ingestion points: Research sources identified in Step 5 (Tier 2 and Tier 3 sources).
- Boundary markers: The 'Skill Security' section explicitly instructs the agent to sandbox its contextual memory to prevent instructions in blogs from exfiltrating data.
- Capability inventory: No tools, scripts, or subprocess permissions are defined in the frontmatter of this skill.
- Sanitization: The 'System Integration Security' section mandates stripping HTML and JavaScript from external research before storage or processing.
Audit Metadata