command-guide
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Analysis of the skill instructions and metadata revealed no malicious patterns. The content is entirely instructional and serves to provide a framework for tool selection within the intended development environment.
- [PROMPT_INJECTION]: No override markers, bypass attempts, or instructions to ignore safety protocols were detected. The 'Auto-Trigger Rules' describe intended operational logic for task transitions rather than attempts to hijack the agent's core behavior.
- [DATA_EXFILTRATION]: There are no network operations (curl, wget, fetch) or attempts to access sensitive file paths (e.g., .ssh, .aws, .env). All references to tools and commands are limited to the internal functionality of the Claude Code platform.
- [COMMAND_EXECUTION]: The skill documents the use of various platform-specific slash commands (e.g., /plan, /compact, /loop). These references are informational and do not involve the execution of arbitrary or dangerous system commands.
- [OBFUSCATION]: The content is written in clear, plain-text markdown and Mermaid syntax. No Base64, zero-width characters, homoglyphs, or other encoding techniques were found.
Audit Metadata