add-feature
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs routine software development tasks such as reading project configuration and generating source code following existing patterns. No malicious behavior or obfuscation was detected.
- [COMMAND_EXECUTION]: The skill includes a step to run
npm run build, which is a standard verification process in JavaScript development to ensure code correctness. - [PROMPT_INJECTION]: The skill exhibits surface area for indirect prompt injection. 1. Ingestion points: Untrusted user input via the
$ARGUMENTSvariable inSKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: File reading (package.json, core files), file writing (Step 3), and command execution (npm run buildin Step 4). 4. Sanitization: Absent. The structural safeguards provided by the step-by-step instructions mitigate the risk associated with this ingestion.
Audit Metadata