superglue-ai

superglue-ai / clawguardian

Public

Security plugin for OpenClaw that detects and filters sensitive data in tool calls.

moltbot moltbot-plugin openclaw openclaw-security openclawplugin
29
1
100% credibility
Found Feb 05, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

ClawGuardian is a security add-on for the OpenClaw AI agent framework that automatically detects and handles sensitive personal information, secret codes, and potentially harmful commands in tool executions.

How It Works

1
🔍 Discover safety for your AI helper

You hear about OpenClaw, a smart AI assistant, and ClawGuardian, its built-in protector against sharing secrets or running risky tasks.

2
🛡️ Add the guardian easily

With one simple step, you turn on ClawGuardian to watch over your AI's actions.

3
⚙️ Choose your protections

You pick what to safeguard, like hiding passwords, personal numbers, or blocking dangerous commands, feeling in full control.

4
💬 Chat and task your AI

You talk to your AI agent, asking it to do everyday jobs like checking files or running commands.

5
Guardian spots something
✅
All clear

No issues found, your task runs smoothly and happily.

⚠️
Risk detected

It hides secrets or asks for your okay on risky steps, keeping things secure.

6
🔒 Enjoy safe results

Secrets stay hidden, dangers are stopped, and you get clean, trustworthy outputs.

🎉 Worry-free AI power

Your AI helper works powerfully and safely, giving you peace of mind every time.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 29 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is clawguardian?

ClawGuardian is a TypeScript security plugin for OpenClaw that scans tool calls for sensitive data like API keys, tokens, cloud credentials, PII (SSNs, credit cards, emails, phones), and destructive commands such as rm -rf or sudo. It blocks, redacts, or requires confirmation on risky inputs and outputs, preventing leaks in AI agent workflows. Install via `openclaw plugins install clawguardian` and tweak actions in your OpenClaw JSON config.

Why is it gaining traction?

Unlike basic github security scanning or alerts, it targets AI tool calls directly—redacting secrets in params, flagging git reset --hard, or prompting for sudo. Configurable severity rules, allowlists for safe tools, and custom regex patterns make it flexible without false positives killing productivity. Devs love the agent-confirm flow, where models retry with a flag after warnings.

Who should use this?

OpenClaw users building agents with exec/bash tools, especially teams handling real API keys or customer data in security github projects. Ideal for security plugin admin roles auditing AI outputs, or devs mimicking github security policy in local repos to avoid accidental PII exposure during calls.

Verdict

Grab it if you're on OpenClaw—solid docs, tests, and MIT license make setup painless despite 20 stars and 1.0% credibility score signaling early maturity. Pairs well with github security actions for full-stack agent safety, but watch for edge cases in custom patterns.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.