research
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a highly structured and restricted research workflow. It mandates that the agent must observe the local tech stack and read existing code before initiating any external research, which prevents the agent from making uninformed recommendations based on irrelevant data.
- [SAFE]: The protocol includes a 'Self-critique' section that acts as a set of logical guardrails, requiring the agent to fail the run if it attempts to suggest recommendations without a source or proceeds to implementation without explicit user consent.
- [SAFE]: External tool usage is limited to documentation resolution and web search via the Firecrawl and Context7 MCP servers. These tools are used within the context of information gathering, and the skill provides clear guidance on establishing a trust hierarchy (e.g., official docs over community blogs) to ensure the integrity of the data processed.
Audit Metadata