awesome-jev-contributor
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONCREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to perform repository checks using commands like
npm ci --ignore-scriptsandnpm run check. It includes specific security instructions to inspect all commands before execution and to use the--ignore-scriptsflag to prevent the execution of potentially malicious lifecycle scripts during dependency installation. - [CREDENTIALS_UNSAFE]: The instructions contain explicit safeguards against credential exposure, directing the agent not to load private
.envfiles and to ensure that ambient credentials are not exposed to the code being reviewed. - [EXTERNAL_DOWNLOADS]: The skill retrieves contribution guidelines, templates, and catalog data from the
AppitStudio/awesome-jevrepository on GitHub. These are legitimate resources owned by the skill's vendor and are used for retrieving the most current repository rules. - [INDIRECT_PROMPT_INJECTION]: As the skill is designed to ingest and analyze external project artifacts (READMEs, code, and tests), it acknowledges the risk of untrusted data. It mitigates this by requiring the agent to record reviewed versions/dates and instructing it to distinguish clearly between source inspection, mocked execution, and live evaluation.
Audit Metadata