source-command-polish
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill definition is purely instructional and focuses on improving project presentation through documentation, metadata updates, and quality checks. It does not contain any executable code, network requests, or attempts to access sensitive system information. Besides, the vendor 'SylphxAI' is recognized as the author, and no suspicious behavior related to third-party resources was found.
- [NO_CODE]: There are no scripts, binaries, or command-line instructions provided within the skill. The execution steps describe high-level tasks for the AI agent to perform using its internal capabilities on project documentation files.
- [INDIRECT_PROMPT_INJECTION]: While the skill involves reading project data (like existing README files or source code) which could theoretically contain malicious instructions, the intended actions (formatting, adding metadata, checking links) are low-risk and restricted to documentation and metadata improvements.
Audit Metadata