claude-code-guide
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): No patterns of prompt injection or system override attempts were detected. The content is purely instructional.\n- [Data Exposure & Exfiltration] (SAFE): The guide mentions configuration files and API keys but uses placeholders (e.g., 'your-key') and includes explicit warnings against sharing sensitive data.\n- [Obfuscation] (SAFE): No encoded, hidden, or deceptive content was found in the documentation.\n- [Indirect Prompt Injection] (SAFE): Although the guide describes tools that can fetch external web content, the skill itself is documentation and does not implement vulnerable data handling logic.\n- [Dynamic Execution] (SAFE): The guide describes the REPL environment as a standard feature for mathematical and data analysis within the Claude Code toolset.
Audit Metadata