Agent Security Hardening AI Agents

Security layers, sandboxing, and defensive plugins that protect AI agents from malicious inputs, unauthorized tool access, and operational vulnerabilities. Does NOT include general application security, DevSecOps platforms, or non-agent-specific threat detection.

There are 181 agent security hardening agents tracked. 5 score above 50 (established tier). The highest-rated is 23blocks-OS/ai-maestro at 68/100 with 525 stars. 2 of the top 10 are actively maintained.

Get all 181 projects as JSON

curl "https://pt-edge.onrender.com/api/v1/datasets/quality?domain=agents&subcategory=agent-security-hardening&limit=20"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.

# Agent Score Tier
1 23blocks-OS/ai-maestro

AI Agent Orchestrator with Skills System - Give AI Agents superpowers:...

68
Established
2 ImKKingshuk/LockKnife

LockKnife: The Ultimate Android Security Research Tool. A unified TUI...

63
Established
3 FlineDev/ContextKit

Claude Code context engineering & planning system for individual AI...

53
Established
4 backbay-labs/clawdstrike

Runtime security enforcement and threat hunting engine for autonomous AI...

53
Established
5 conorluddy/ios-simulator-skill

An IOS Simulator Skill for ClaudeCode. Use it to optimise Claude's ability...

53
Established
6 mishabar410/PolicyShield

Declarative firewall for OpenClaw tool calls.

48
Emerging
7 Community-Access/accessibility-agents

Accessibility review agents for Claude Code, GitHub Copilot, and Claude...

48
Emerging
8 peg/rampart

Open-source firewall for AI agents. Policy engine that controls what...

47
Emerging
9 adversa-ai/secureclaw

SecureClaw - Security Plugin and Skill for OpenClaw OWASP-Aligned

44
Emerging
10 secnova-ai/ClawdSecbot

ClawdSecbot is a professional AI Bot security protection solution, providing...

44
Emerging
11 SleuthCo/clawshield-public

Security proxy for AI agents. Scans every message for prompt injection, PII,...

43
Emerging
12 fullstackcrew-alpha/privacy-mask

Automatically redacts sensitive data in screenshots before sending to AI agents

42
Emerging
13 ameyalambat128/swiftui-skills

Agent skills for SwiftUI, built from Apple's Xcode AI documentation.

42
Emerging
14 jnMetaCode/shellward

首个中文 OpenClaw 安全插件 | 身份证号/手机号/银行卡号自动保护 | 中文注入攻击检测 | 数据外泄链拦截 |...

41
Emerging
15 DinoMorphica/safeclaw

Security Dashboard for OpenClaw AI Agents - intercept, monitor, and control...

39
Emerging
16 dativo-io/talon

Intercept every AI call you care about. Scan for PII, enforce budgets,...

38
Emerging
17 Bin-Huang/google-search-console-cli

Google Search Console CLI designed for AI agents. Wraps the official Search...

38
Emerging
18 NinoSkopac/openclaw-secure-kit

Secure-by-default OpenClaw on Ubuntu, with a verifiable security report

38
Emerging
19 Parad0x-Labs/liquefy-openclaw-integration

Liquefy is a local-first OpenClaw vault system for packing, verifying,...

37
Emerging
20 xinxin7/claw-shield

The governance layer for AI agents — monitor reasoning, audit tool calls,...

37
Emerging
21 cybrlab-ai/preclick-mcp

Safe Agentic Browsing. Assess target URLs for potential threats and...

36
Emerging
22 X-Scale-AI/grits-audit

Security scanner and auto-fixer for OpenClaw and NemoClaw AI agents. Scan,...

35
Emerging
23 rad-security/clawkeeper

Open-source security scanner for AI agent hosts. Audits macOS and Linux...

35
Emerging
24 genecell/stato

Capture, validate, and transfer AI agent expertise. Across sessions,...

34
Emerging
25 ClawTower/ClawTower

OS-level runtime security for AI agents. Tamper-proof monitoring,...

34
Emerging
26 rbarooah/bike-tool

An agent-focused Swift CLI for safely reading and editing Bike.app .bike...

34
Emerging
27 WalterSumbon/minicode-sdk

A Python SDK for building AI agents with LLM, tools, skills, and MCP...

34
Emerging
28 jfr992/openclaw-sentinel

🦀 openclaw sentinel — Real-time security monitoring for openclaw agents

34
Emerging
29 AlexAnys/ai-sub-auth

Reuse your AI subscriptions. One module, every provider. OAuth PKCE for...

33
Emerging
30 KreerC/ACCESSIBILITY.md

SKILL.md for web accessibility by real accessibility experts. Make sure your...

33
Emerging
31 merciagents/riphook

Deterministic security layer for Openclaw(Clawdbot), Cursor and Claude Code....

32
Emerging
32 4ier/a11y-bridge

16KB Android Accessibility Bridge for AI agents. Read UI trees in 50ms,...

32
Emerging
33 cybrlab-ai/urlcheck-openclaw

URLCheck MCP security scanner plugin for OpenClaw. Analyzes URLs for...

32
Emerging
34 Anecdotes-Yair/trust-my-agent-ai

Security & Compliance bodyguard for OpenClaw agents

32
Emerging
35 gygantskiyMatilyock/ios-developer-agents

AI Agents and Skills for iOS Developers

31
Emerging
36 M64GitHub/clawgate

Zero-trust capability proxy for AI agents. Scoped access to files, git, and...

30
Emerging
37 Eversec-ai/AIGov-Insight-Web

Eversec AIGov-Insight Web

29
Experimental
38 raintree-technology/hig-doctor

Apple HIG audit CLI + 14 agent skills for AI coding assistants — score any...

29
Experimental
39 Techopolis/swift-agents

A team of specialized Swift agents for Claude Code. Swift 6.2 concurrency,...

27
Experimental
40 jetywolf/claw-diary

Audit log + guard for AI agents. Passive logging, human-in-the-loop approval...

26
Experimental
41 raiph-ai/fireclaw

🛡️ A firewall for your agent's brain. Open-source security proxy that...

26
Experimental
42 cdot65/prisma-airs-plugin-openclaw

OpenClaw plugin for Prisma AIRS from Palo Alto Networks

26
Experimental
43 artemnovichkov/xcode-tools-docs

Missing Xcode Tools Documentation

25
Experimental
44 mattijsmoens/openclaw-sovereign-shield

SovereignShield native security plugin for OpenClaw. Deterministically...

24
Experimental
45 24kchengYe/desktop-controller-skill

AI Computer Use for Claude Code — The open-source alternative to OpenAI...

24
Experimental
46 Marways7/cua_desktop_operator_skill

MCP skill that lets any AI agent operate a Windows desktop — clone-ready,...

24
Experimental
47 agenticpoa/jean-claw-van-damme

Authorization gatekeeper for OpenClaw agents. Scoped grants, time-bound...

23
Experimental
48 wd041216-bit/openclaw-github-repo-commander

GitHub repository management skill powered by the 7-Stage Super Workflow —...

23
Experimental
49 Soulreaper1188/openclaw-detect

🔍 Detect OpenClaw on managed devices with lightweight scripts for macOS,...

23
Experimental
50 Neros0/defillama-watchdog

Proactive DeFi sentinel for OpenClaw. Monitors TVL drops, protocol hacks,...

23
Experimental
51 joergmichno/clawguard

🛡️ Security Scanner for AI Agents — 42+ detection patterns across 5...

23
Experimental
52 Dorukuz/Claude-Code-IOS-App-Studios

Claude Code + Cursor template: structured AI studio for Swift/SwiftUI iOS...

23
Experimental
53 NeuZhou/clawguard

🛡️ ClawGuard — AI Agent Immune System. 285+ threat patterns, PII sanitizer,...

23
Experimental
54 mladjan/Sclawhub

🦞🛡️ Security scanner for OpenClaw AI agent skills

23
Experimental
55 DmiyDing/watchdog-shrimp

OpenClaw execution governance skill that reduces low-risk confirmation noise...

23
Experimental
56 Goki602/ClawGuard

AI Agent Memory — Fewer prompts, smarter decisions. Remembers your trust...

23
Experimental
57 sundarsub/execwall

OpenClaw Execution Firewall - Seccomp-locked AI agent sandbox with...

23
Experimental
58 stronghold-hq/stronghold

A CLI-driven security proxy that scans every HTTP request for threats using...

23
Experimental
59 Marways7/cua_desktop_operator_cli_skill

CLI skill that lets any AI agent operate a Windows desktop — clone-ready,...

23
Experimental
60 glenwille12/ios-developer-agents

🔍 Validate, audit, and enhance your iOS apps with AI agents to ensure...

23
Experimental
61 sanna-ai/sanna-openclaw

Constitution enforcement and cryptographic receipts for OpenClaw agents....

23
Experimental
62 NeoSkillFactory/openclaw-sandbox-manager

Creates and manages a sandboxed environment for OpenClaw, isolating system...

22
Experimental
63 InnerWarden/openclaw-integration

InnerWarden security skill for OpenClaw — validates commands, monitors...

22
Experimental
64 aceclaw826/clawdit

🦞🔍 Belief systems auditor for OpenClaw agents. A separate agent that...

22
Experimental
65 Meow-AIs/BlackCat

Your DevSecOps engineer and Solution Architect — in a 15MB binary. Scan...

22
Experimental
66 lawcontinue/aegis-skill

Aegis - OpenClaw智能优化插件。提供模型选择建议、Prompt优化、成本统计和质量评估功能。

22
Experimental
67 ThankNIXlater/memory-guard

Agent memory integrity monitoring. Detect injection, tampering, drift, and...

22
Experimental
68 ayuday/gstack

AI Skills: Claude Code setup: 15 opinionated tools that serve as CEO,...

22
Experimental
69 wadim/haveibeenclawned

Security audit for OpenClaw agents — 72 checks, one command

22
Experimental
70 Erikote04/Swift-API-Design-Guidelines-Agent-Skill

Swift API Design Guidelines Agent Skill for AI coding tools that support the...

22
Experimental
71 ivan-magda/swift-claude-code

A Swift reimplementation of a Claude Code-style coding agent, built stage by...

22
Experimental
72 NeoSkillFactory/sandbox-guard

Automatically detects and mitigates sandbox escape attempts to prevent...

22
Experimental
73 joergmichno/clawguard-shield

ClawGuard Shield — REST API for AI agent security scanning. 10 endpoints,...

22
Experimental
74 shane9coy/katana-agent

⚡ Drop your AI agent into any project. Centralized Obsidian memory across...

22
Experimental
75 aa-on-ai/clawbotomy

Behavioral stress tests for AI agents. 12 tests, 6 dimensions, trust scores...

22
Experimental
76 Resolver-TNG/ogas-openclaw

OGAS - OpenClaw Guard Agent Security System | Memory integrity & cron...

22
Experimental
77 ClawSecure/clawsecure-openclaw-security

Free OpenClaw security scanner. 2,890+ agents audited. 3-Layer Audit...

22
Experimental
78 jamebobob/openclaw-agent-privacy

Layered privacy framework for multi-agent OpenClaw deployments. N:M memory...

22
Experimental
79 mackeh/AegisClaw

Secure-by-default runtime and security envelope for OpenClaw AI agents....

22
Experimental
80 dazeb/openclaw-sentinel

OpenClaw Sentinel - Security for OpenClaw agents

22
Experimental
81 jamebobob/openclaw-privacy-protocol

Teach your AI agent what's private. Three concepts, not a thousand rules.

22
Experimental
82 medhanshy27-cyber/cua_desktop_operator_skill

Deliver a local, Windows-based execution layer for MCP-capable agents with...

22
Experimental
83 no-hup/unjank

Agent skills for tracking and fixing screen rendering performance in...

22
Experimental
84 shahrilarif00/openclaw-guardian

Protect AI agent fleets by monitoring and hardening their operation to...

22
Experimental
85 griffithfly/OpenClaw-filesafe

Zero-friction file protection for OpenClaw workspaces. One command tool to...

22
Experimental
86 ekswathi/SafeClaw

🔒 Securely manage AI agents with hardware-isolated execution and message...

22
Experimental
87 forgestudi0s/clawshield

🛡️ Enhance AI agent security with a lightweight, zero-config layer offering...

22
Experimental
88 razashariff/agentsign-openclaw

AgentSign zero trust middleware for OpenClaw and NemoClaw -- cryptographic...

22
Experimental
89 drakerfire98/openclaw-skill-security-practices

9-layer defense-in-depth security architecture for OpenClaw skill loading

22
Experimental
90 DafaSya/ios-developer-agents

🔍 Validate, audit, and improve your iOS apps with AI agents to ensure...

22
Experimental
91 amirhosseinkab/swift-agent-team

Provide specialized Swift agents for Claude Code that support Swift 6.2...

22
Experimental
92 Vikas-Kumar-Sahu/openclaw-security-practice-guide

Provide security guidelines for High-Privilege Autonomous AI Agents using...

22
Experimental
93 jeremieb/swift-unit-test-instructions

A reusable framework of Claude Code instructions for Swift, SwiftUI, and...

22
Experimental
94 PleaseChooseUsername/aegis-openclaw-skill

AEGIS — Automated Emergency Geopolitical Intelligence System. An OpenClaw...

22
Experimental
95 syncropic/syncropel-cli

spl — the Syncropel CLI and local developer registry. Full governance stack...

22
Experimental
96 beav3r-ai/openclawXBeav3r

Approve Openclaw auth prompts on the go using Beav3r.

22
Experimental
97 Ken-Chy129/clawhub-scanner

Implements the same security scanning logic as ClawHub's built-in Security...

22
Experimental
98 isaackaara/openclaw-api-proxy

Zero-knowledge API proxy for AI agents. Keys live on your server, agents...

22
Experimental
99 liu971227-sys/openclaw-audit

OpenClaw security audit and hardening CLI for AI agents, LLM agent...

22
Experimental
100 humancto/punch

The Agent Combat System — Deploy conscious AI agent squads. 103 bundled...

22
Experimental
101 jamebobob/openclaw-privacy-guardrail

OpenClaw plugin that blocks direct writes to public paths. Forces staging...

22
Experimental
102 Protocol-zero-0/openclaw-security-radar

Security checklists, review templates, and builder guidance for the OpenClaw...

22
Experimental
103 jamebobob/openclaw-read-guardrail

OpenClaw plugin that prevents one agent from reading another agent's...

22
Experimental
104 jerisadeumai/zero-shield-cli

Zero-Shield CLI is a security-hardened AI Copilot for AWS, built on a...

22
Experimental
105 yaseenkadlemakki/clawaudit

Runtime security intelligence platform for OpenClaw AI agents — skill...

22
Experimental
106 cybrlab-ai/preclick-openclaw

An intent + security layer for agent browsing that verifies each destination...

22
Experimental
107 royyamamoto-rgb/shellarmor

Docker-native security hardening for OpenClaw gateways. No Kubernetes needed.

22
Experimental
108 badass-courses/a11y-agent

Accessibility expertise for AI coding agents.

22
Experimental
109 Yelsmeky/apple-diagnostics-ai-agent-skill

AI agent and reusable skill for diagnosing macOS and Apple ecosystem...

22
Experimental
110 RADobson/e8cr-squad

Autonomous Essential Eight ML2 compliance bots — open-source OpenClaw skills...

22
Experimental
111 lhhiep2204/Apple-Agent-Builder-Kit

Apple-first Copilot agent builder kit for iOS, iPadOS, macOS, watchOS, tvOS,...

22
Experimental
112 Kennethhuang67/openclaw-iso27001

ISO 27001 security compliance scanner for OpenClaw AI agents

22
Experimental
113 hkai614119-star/Claw-Ghost

Trusted private execution demo kit for Claw / Onchain OS, focused on intent...

22
Experimental
114 AndreaGriffiths11/claw-relay

A trust layer between AI agents and your real browser. Token auth, scoped...

22
Experimental
115 F4bioo/berry-shield

Security guardrails plugin for OpenClaw: layered runtime policy enforcement...

22
Experimental
116 jav-ed/web-audit

CLI accessibility audit tool — axe-core checks in dist (static HTML) or dev...

22
Experimental
117 cyberindranil/claw-diary

Enable secure, auditable control and collaboration for AI agents with a...

22
Experimental
118 saurabhdave/aiagents

AI‑driven skills for Apple platform engineers (accessibility, performance,...

21
Experimental
119 jugaad-lab/clawguard

Security blacklist protecting AI agents from malicious skills, scams, and...

21
Experimental
120 Patrickschell609/ghostclaw

The AI agent that runs code in a kernel sandbox. Ed25519 signing. Landlock +...

21
Experimental
121 topazyo/openclaw-security-playbook

Production-ready security playbook for AI agent deployments. Addresses...

21
Experimental
122 clawhatch/clawhatch

Security scanner for OpenClaw AI agent configurations. 128 checks, MIT...

21
Experimental
123 anoopkansupada/clawscan

AI Agent Security Scanner - Detect exposed API keys, misconfigurations, and...

20
Experimental
124 Clavion-xyz/clavion

Secure crypto runtime for autonomous agents — key isolation, policy...

20
Experimental
125 lennystepn-hue/clawshield

🛡️ Security Layer for AI Agents. One binary. Zero config. 50+ checks. Linux...

20
Experimental
126 tobiassved/openclaw-best-practices

Comprehensive security and best practices guide for deploying AI agents with...

20
Experimental
127 albertdobmeyer/openclaw-vault

Hardened container sandbox for OpenClaw agents — API keys never enter the container

20
Experimental
128 albertdobmeyer/clawhub-forge

Offline-first skill development workbench with 87-pattern security scanner

20
Experimental
129 darbybailey/kokeshi-beanstalk

Kokeshi Beanstalk - Security hardening runtime guardian for AI agents

20
Experimental
130 hacksurvivor/i18n-localizer

AI-powered i18n/localization skill for web, mobile, and SwiftUI projects.

20
Experimental
131 AlphaSquadTech/ios-dev

Agent Skill for autonomous iOS app development — build, run, screenshot,...

20
Experimental
132 goheesheng/x402guard

x402-powered security auditing for AI agent skills †x

19
Experimental
133 5WARM-AI/swarm-leak-detector

Free credential leak detection for OpenClaw agents. Scans for 20+ API key...

19
Experimental
134 guillempuche/ai-agent-a11y-accessibility-reviewer

Use this agent to review code for accessibility (a11y) compliance. Use after...

19
Experimental
135 avp-protocol/zeroclaw-avp

ZeroClaw SecretBackend using AVP

19
Experimental
136 soarealin/vettai

Vet your AI skills before they vet you. Security scanner for AI agents — we...

19
Experimental
137 fluent-official/accessibility-tools

Fluent's suite of accessibility wrappers for your applications. Fluent is an...

19
Experimental
138 liangfeng-hu/openclaw-flight-recorder-verified

🦞 OpenClaw Flight Recorder PoC · local-only sidecar · RFC-001 JSONL →...

19
Experimental
139 virtaava/security-audit

Fail-closed security auditing for OpenClaw/ClawHub skills & repos...

19
Experimental
140 ca7ai/openclaw-audit

A lightweight security configuration scanner and hardening tool for OpenClaw...

19
Experimental
141 tomstetson/declaw

Security-hardened fork of OpenClaw — defense-in-depth for AI agents (secrets...

19
Experimental
142 joshivignesh/accessibility-agent

An intelligent accessibility checking and fixing agent for VB.NET, React,...

19
Experimental
143 P3tr4-sway/moltbot-security-hardening-guide

��️ The Ultimate 13-Layer Security Hardening Guide for Moltbot/OpenClaw AI...

19
Experimental
144 agentguard-ai/tealtiger-python-prod

TealTiger Python SDK - Drop-in security and cost tracking for OpenAI,...

19
Experimental
145 dalbit-mir/browser-use-skill

Claude Code Skill wrapping official browser-use library. Two modes: Direct...

17
Experimental
146 X-Scale-AI/openclaw-security

Harden your OpenClaw and NemoClaw AI agent installations -- see grits-audit

17
Experimental
147 lauty1505/clawguard

🔍 Monitor AI activities and enhance security with ClawGuard, the real-time...

16
Experimental
148 ZSeven-W/skill-cli

Cross-platform CLI for creating, validating, and converting AI agent skills....

16
Experimental
149 rubendario921/openclaw-prism

Protect OpenClaw gateways by injecting runtime defenses to block prompt...

16
Experimental
150 mohahasan/ios-agentic-skills

🔍 Discover and utilize agentic iOS/watchOS audit skills and playbooks for...

16
Experimental
151 JuanAtLarge/openclaw-safe

Audit, harden, and monitor your OpenClaw install — auto-quarantine malicious...

15
Experimental
152 oktsec/oktsec-openclaw

oktsec plugin for OpenClaw. Runtime security for AI agent tool calls. 188...

15
Experimental
153 Guent26/xcode-tools-docs

🛠️ Enable AI assistants to manage Xcode workspaces with Xcode Tools,...

15
Experimental
154 stephenlzc/claw-gatekeeper

A safety brake for OpenClaw with session-aware risk management

14
Experimental
155 mj-deving/openclaw-hardened

Security-first OpenClaw deployment — hardened config, monitoring scripts,...

14
Experimental
156 SunrisesIllNeverSee/moses-claw-gov

MO§ES™ Constitutional Governance Harness — modes, postures, roles, SHA-256...

14
Experimental
157 nutstrut/verified-capability-evolver

Controlled self-improvement system for AI agents with verification,...

14
Experimental
158 slvssb/skills-ui

Native macOS app for managing Vercel Labs Skills — browse, install, and...

14
Experimental
159 Unawakened-landlord758/ClawGuard

Protect OpenClaw by blocking harmful actions, scanning skills, and...

14
Experimental
160 iamhitarth/secure-my-claw

Security hardening guide for OpenClaw AI agents. Covers secrets management,...

14
Experimental
161 Nishad710/desktop-controller-skill

Enable AI-driven control of Windows apps with native desktop and web...

14
Experimental
162 helper618/accessibility-agents

Provide automated accessibility checks to support developers in identifying...

14
Experimental
163 peeweeh/mantou

Local-first security posture scanner for OpenClaw agents. 69 rules. Zero telemetry.

14
Experimental
164 ahmadqatato2-maker/openclaw_vulnerabilities_and_solutions

Document vulnerabilities and solutions for deploying OpenClaw on VPS with...

14
Experimental
165 FrigateCaptain/openclaw_vulnerabilities_and_solutions

> OpenClaw security audit and hardened deployment guide — known...

13
Experimental
166 StarkTechIndustries/stark-shield

A living immune system for OpenClaw AI agents. Three-part security...

12
Experimental
167 pranit-gandhi/agent-locker

Agent Locker (CloseClaw) — Sandbox for AI Agents

12
Experimental
168 thefooliman/ClawShield

A Visual Security Sentry for AI Agents. Stop accidental clicks before they happen.

12
Experimental
169 mangiapanejohn-dev/Resonix-Skill

A production-ready AI Agent operating system with multi-layer memory,...

12
Experimental
170 dineshuday/openclaw-utm-sandbox

A secure, sandboxed implementation of OpenClaw running in a macOS UTM...

11
Experimental
171 getmilodev/milo-shield

OpenClaw security hardening and optimization tools by Milo

11
Experimental
172 getmilodev/milo-scan

Free OpenClaw security scanner. Run: npx milo-scan

11
Experimental
173 ClawrenceSystems/openclaw-hardening

OpenClaw security hardening guide — safe setup, integrations, and best practices

11
Experimental
174 virtaava/openclaw-hardener

Harden OpenClaw: run openclaw security audit, reduce prompt-injection/exfil...

11
Experimental
175 aceman23/TheGoLiveChecklistforCoders

This repository contains a standardized checklist to ensure that web...

11
Experimental
176 mastrophot/near-phishing-detector

Security-focused OpenClaw skill for detecting phishing URLs and contracts on NEAR.

11
Experimental
177 PiQrypt/piqrypt-openclaw-integration

Verifiable Memory bridge for OpenClaw_tamper-proof audit trail for autonomous agents

11
Experimental
178 arc-claw-bot/clawdefender

Input security scanner for AI agents. Detects and prevents prompt injection,...

11
Experimental
179 arc-claw-bot/arc-sentinel

Security monitoring toolkit for OpenClaw agents. Automated security checks,...

11
Experimental
180 getmilodev/milo-watch-pro

Advanced OpenClaw security monitoring with trends, cost analysis, and...

11
Experimental
181 getmilodev/milo-watch

Daily automated security & health monitoring for OpenClaw deployments. Free skill.

11
Experimental