research-verify

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill defines a structured workflow for research verification. It involves reading project-specific files (e.g., proposal.md, experiment-plan.md) and generating verification reports or amendments.- [NO_CODE]: The skill contains no executable scripts or command-line operations, significantly reducing the attack surface.- [DATA_EXPOSURE_AND_EXFILTRATION]: All file operations are restricted to local research directories (research/proposals/, research/idea-amendments/). No network operations or external data transmissions are present.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 07:10 AM