code-splitting-helper
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill is permitted to use the
BashandWritetools. While appropriate for frontend development tasks, these tools represent a high-capability surface that could be exploited if the agent executes code derived from untrusted user input. - [PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection. 1. Ingestion points: User requests and code provided for splitting analysis. 2. Boundary markers: None present. 3. Capability inventory: Bash, Write, Edit, Grep. 4. Sanitization: No explicit validation or escaping of external content is defined.
- [DATA_EXFILTRATION] (SAFE): No hardcoded credentials, sensitive file access (e.g., SSH keys), or suspicious network operations were found.
- [EXTERNAL_DOWNLOADS] (SAFE): No remote scripts, unverified packages, or automated downloads were detected in the skill definition.
Audit Metadata