skill-evaluation
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The documentation provides manual installation commands for complementary skills using the
npx skills addtool. These commands reference well-known and reputable organizations including Anthropic's official skills repository and specialized academic tools from established contributors. These references are documented as standard extension of the research workflow. - [COMMAND_EXECUTION]: The skill references local validation scripts (
python3 scripts/validate_skills.py) and standard testing frameworks (pytest). These are used for structural verification of the research package and represent best practices for skill development rather than malicious execution. - [SAFE]: The skill uses
references/files for policy documentation and defines clear boundary markers for how an agent should evaluate research tasks. The YAML configuration is consistent with the skill's stated purpose of auditing and refining academic workflows.
Audit Metadata