Unsafe Checker
This tool analyzes prompts for potentially harmful or dangerous instructions, safeguarding against misuse and promoting responsible AI interaction.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add actionbook-unsafe-checker npx -- -y @trustedskills/actionbook-unsafe-checker
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"actionbook-unsafe-checker": {
"command": "npx",
"args": [
"-y",
"@trustedskills/actionbook-unsafe-checker"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
The unsafe-checker skill analyzes Rust code to identify potential safety violations and unstable features. It helps developers ensure their projects adhere to safe coding practices by flagging risky patterns before they cause runtime errors.
When to use it
- Reviewing third-party crates for hidden unsafe blocks or deprecated APIs.
- Validating complex logic in performance-critical modules where manual auditing is error-prone.
- Integrating into CI/CD pipelines to automatically block unsafe code commits.
- Auditing legacy Rust codebases before refactoring them to modern standards.
Key capabilities
- Detects usage of
unsafeblocks and warns about their implications. - Identifies unstable features marked with
#![feature(...)]. - Highlights deprecated functions or types that may break in future versions.
- Provides context-aware suggestions for safer alternatives.
Example prompts
- "Scan this Rust module for any unsafe code and explain why it's risky."
- "Check if this crate uses any unstable features that might cause compilation errors later."
- "Analyze the following function for deprecated API usage and suggest replacements."
Tips & gotchas
This tool is designed specifically for Rust projects; applying it to other languages will yield no results. While it catches many issues, it should complement—not replace—manual code reviews by experienced developers.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.