taste-skill
Warn
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions facilitate the download of external code from an untrusted GitHub repository (Leonxlnx/taste-skill) via npx skills add or /plugin commands. \n- [REMOTE_CODE_EXECUTION]: Installation of the full skill enables the execution of Node.js scripts and browser-based automation tools originating from the untrusted third-party repository. \n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by fetching assets (pictures) from external web collections and analyzing web page content/screenshots. This surface could be exploited if external content contains hidden instructions. \n
- Ingestion points: External image collections, browser inspector data, and screenshots of processed web pages. \n
- Boundary markers: No explicit markers or 'ignore' instructions are defined for the ingested data. \n
- Capability inventory: Execution of shell commands, Node.js scripts, browser automation, and file system writes to the .taste-skill/ directory. \n
- Sanitization: No evidence of sanitization or validation of the fetched external assets or inspector output.
Audit Metadata