feedback
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill executes
npx -y @tiny-fish/cli@latest doctorto gather diagnostic information. This fetches and runs the latest version of the vendor's CLI tool from the public npm registry. - [COMMAND_EXECUTION]: The skill utilizes shell commands like
npxand the GitHub CLI (gh). It demonstrates high security awareness by providing specific instructions to avoid shell injection vulnerabilities, such as using file-based inputs (--body-file) and avoidingevalor direct interpolation of untrusted user text into command strings. - [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data from user feedback and diagnostic reports. It implements robust defenses by requiring a preview gate and explicit user consent before any data is transmitted.
- [DATA_EXFILTRATION]: While the skill transmits data to an external service (GitHub), the destination is hardcoded to the vendor's own repository. The diagnostic tool used (
tinyfish doctor) is described as having built-in redaction to ensure no sensitive credentials or configuration are included in the report.
Audit Metadata