user-research
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access to sensitive system resources were detected. The skill adheres to its stated purpose of assisting with user research and data analysis.- [PROMPT_INJECTION]: The skill processes untrusted external data in the form of user interview transcripts and survey results, which creates a theoretical surface for indirect prompt injection. This is mitigated by the 'Anti-Hallucination Rules' and 'Quality Gate' requirements that mandate strict source validation, separation of facts from inferences, and confidence ratings for all insights.
- Ingestion points: Reads project-level market data from
docs/product/.ompm/market-analysis.jsonand processes user-provided research inputs. - Boundary markers: Explicit 'Anti-Hallucination Rules' are defined to ensure data is validated against real sources before being incorporated into the agent's context.
- Capability inventory: Performs filesystem writes to the project directory (
docs/product/.ompm/user-research.json) and utilizes web search capabilities to verify industry trends. - Sanitization: No specific string sanitization is used, but the skill requires structured JSON output and source-backed verification for all findings.- [EXTERNAL_DOWNLOADS]: The skill involves web search operations to fetch and verify industry benchmarks and statistical trends. This behavior is essential to its primary function and includes strict rules requiring full URL sources for every data point gathered.
Audit Metadata