Zen-Open-Source

Zen-Open-Source / AgentArmor

Public

Security scanning for AI Agents

12
1
100% credibility
Found Mar 30, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

AgentArmor is a security scanner that audits AI agent setups like OpenClaw for vulnerabilities and misconfigurations, assigns a score out of 100, and offers automated fixes.

How It Works

1
🔍 Discover AgentArmor

While setting up your AI helper, you learn about a friendly security checker that keeps it safe from risks.

2
📥 Get the checker

You download the simple all-in-one app that runs anywhere without extra setup.

3
🧐 Check your AI setup

You run the checker on your AI helper's files, and it quietly reviews everything for safety issues.

4
📊 See your safety score

A clear score out of 100 pops up, showing exactly how protected your AI is, with tips on weak spots.

5
Choose your next move
🛠️
Fix issues

The app safely applies fixes, backs up your setup, and boosts your score.

✅
All good

You relax knowing your AI is already hardened against common dangers.

🛡️ Secure AI ready

Your AI helper is now fully protected, giving you peace of mind every time you use it.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 12 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 AgentArmor?

AgentArmor is a Rust-built security scanner for AI agents that audits your local installation—currently OpenClaw—against official security docs, delivering a score out of 100 across 30 checks in categories like auth, file perms, network exposure, and secrets. Run `agent-armor scan` for a quick CLI report with fixes, or `agent-armor fix` to auto-remediate issues like weak tokens or bad Docker flags. It outputs human-readable summaries, JSON for CI, and even flags CVEs, acting as security scanning software for agent configs.

Why is it gaining traction?

Unlike generic github security scanning tools or GitHub security advisories, it targets AI agent risks like prompt injection, sandbox escapes, and tool auth—stuff security github copilot overlooks in local setups. The auto-fixer with dry-run and backups hooks devs needing fast hardening without deep docs dives, plus pluggable design promises more agents soon. Single-binary deploy beats security scanning gitlab bloat for quick audits.

Who should use this?

OpenClaw users securing desktop AI agents for production chats or tools. AI ops engineers scanning agentarmor installs before github actions deploys. Devs building security github projects who want agent-specific checks beyond basic security scanning tools for code.

Verdict

Grab it via `cargo install` if you're on OpenClaw—solid for baseline audits despite 12 stars and 1.0% credibility signaling early days with thin tests. Pair with official docs for maturity; skip for non-OpenClaw agents until plugins expand.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.