babel-configuration-for-nativewind
Warn
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [Indirect Prompt Injection] (MEDIUM): The skill possesses high-privilege write/edit capabilities while ingesting untrusted code files. 1. Ingestion points: External code and 'babel.config.js' files. 2. Boundary markers: Absent. 3. Capability inventory: 'Read', 'Write', and 'Edit' tools are active. 4. Sanitization: No sanitization of ingested content before processing or modification.
- [Command Execution] (LOW): The 'Memory Protocol' section (line 55) specifies a direct shell command 'cat .claude/context/memory/learnings.md' for reading local files, bypassing safer tool-based alternatives.
Audit Metadata