1k-patching-native-modules
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Command Execution] (SAFE): The skill utilizes
rm -rfspecifically for cleaning build artifacts within thenode_modulesdirectory. This is a routine development task and does not target sensitive system paths or user data. - [External Downloads] (SAFE): The instructions utilize
npx patch-package, which is a well-known and trusted utility in the JavaScript ecosystem. The packages referenced (e.g.,expo-image,react-native) are high-trust, standard libraries. - [Indirect Prompt Injection] (LOW): The skill identifies a workflow for analyzing external crash logs, which constitutes an ingestion point for untrusted data. Evidence: 1. Ingestion points: Step 1 (Analyze Crash Log). 2. Boundary markers: Absent. 3. Capability inventory:
ls,rm,npx,git,gh. 4. Sanitization: Absent. The risk is minimized as the skill is primarily instructional and has model invocation disabled.
Audit Metadata