ln-11-opportunity-evaluator
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill evaluates product opportunities by ingesting untrusted data from external sources such as web research, marketplace data, online communities, and competitor pages. This creates a potential surface for indirect prompt injection where malicious instructions embedded in those external sources could attempt to influence the agent's behavior.
- Ingestion points: The skill (specifically in the Tool Routing and Checklist sections) directs the agent to gather data from external market claims, reviews, communities, and competitor websites.
- Boundary markers: The instructions mandate that the agent should 'read selectively' and 'summarize bounded results', which provides some logical separation but not a technical boundary.
- Capability inventory: The skill contains a strong 'read-only' constraint, explicitly forbidding the creation of files, roadmaps, implementation plans, or outreach campaigns, which significantly limits the impact of potential injections.
- Sanitization: There is no evidence of programmatic sanitization or escaping of the ingested content; however, the skill's mandatory evidence classification system (MEASURED, REPORTED, ESTIMATED, etc.) requires the agent to verify the quality and source of information.
Audit Metadata