researcher
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's instructions emphasize verification and source citation when researching third-party code.
- It requires the agent to cite specific URLs and commit SHAs when available.
- It mandates marking unverified claims as
[unverified]. - It explicitly forbids fabricating links, version numbers, or API signatures.
- [SAFE]: The skill implements a "Capability Gate" that restricts agent behavior based on the actual tools provided by the environment, preventing it from assuming unauthorized filesystem or network access beyond what is explicitly granted.
Audit Metadata