dcg
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or behaviors were identified during the analysis. All detected patterns are consistent with the tool's primary purpose as a security guard for AI agents.
- [EXTERNAL_DOWNLOADS]: The skill provides installation and uninstallation scripts hosted on the author's official GitHub repository (github.com/Dicklesworthstone/...). It also includes standard Rust toolchain setup instructions via sh.rustup.rs. These resources are used for the intended purpose of setting up the security tool.
- [PROMPT_INJECTION]: The file 'AGENTS.md' contains instructional directives for AI agents contributing to the development of the project, including a directive ('RULE 0') to follow the owner's commands. These instructions are confined to the development workflow of the tool and do not represent an attempt to bypass user safety filters or hijack agent behavior in general use.
- [COMMAND_EXECUTION]: The codebase, test suites, and documentation contain numerous examples of destructive shell commands (e.g., 'rm -rf /', 'git reset --hard'). These are used exclusively as test fixtures to verify the tool's blocking logic and as documentation to educate users on protected patterns.
Audit Metadata