claude-code-guide
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill mentions the phrase 'forget previous instructions' in the context of a debugging guide for users. This is documented as a manual intervention step for users to resolve agent confusion, not a malicious attempt by the skill author to override the underlying system prompt.
- [COMMAND_EXECUTION]: The instructions include templates for standard development commands such as 'npm run dev' and 'npm test', and recommend using common CLI tools like 'grep' and 'find'. These references are purely illustrative for project workflows and do not involve unauthorized or hidden command execution.
- [DATA_EXFILTRATION]: No network operations or sensitive data access patterns were identified. The single external URL points to a legitimate public GitHub repository, which is a well-known service for documentation.
Audit Metadata