react-19
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): The content consists of standard coding instructions. There are no patterns found that attempt to override system prompts, bypass safety filters, or extract initial instructions.
- [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths (e.g., .ssh, .env), or exfiltration commands were detected. The use of WebFetch/WebSearch in metadata is consistent with a developer tool used for research.
- [Unverifiable Dependencies & Remote Code Execution] (SAFE): No package managers (npm, pip) or remote code execution patterns (curl | bash) were identified in the provided file.
- [Obfuscation] (SAFE): No Base64, zero-width characters, homoglyphs, or encoded strings were found.
- [Indirect Prompt Injection] (SAFE):
- Ingestion points: Processes local .tsx files as specified in the description.
- Boundary markers: Not explicitly defined in this instructional file.
- Capability inventory: Includes access to Bash, WebFetch, and Write tools.
- Sanitization: Not applicable as the skill does not contain executable code, only documentation patterns. No malicious surface for injection was observed.
Audit Metadata