cisco-ai-defense

cisco-ai-defense / defenseclaw

Public

Security Governance for Agentic AI

167
14
100% credibility
Found Mar 28, 2026 at 167 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

DefenseClaw is an enterprise security layer that scans and governs AI agents on OpenClaw by blocking malicious skills, MCP servers, plugins, and runtime threats like prompt injection or data leaks.

How It Works

1
🔍 Discover secure AI helpers

You find DefenseClaw, the safety shield for powerful AI agents that can run code and access the web.

2
📦 Add protection with one click

Download and launch DefenseClaw—it connects automatically to your AI setup without hassle.

3
🛡️ Turn on real-time guards

Enable the guardrail to watch every message and action, blocking dangers before they happen.

4
🔎 Check your AI tools

Scan installed helpers for hidden risks like secret leaks or sneaky commands.

5
🚨 See and fix alerts

Review the dashboard for issues, block risky tools, and keep everything safe.

✅ Enjoy secure AI magic

Your AI agents now work confidently, protected from threats with full activity logs.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 167 to 167 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 defenseclaw?

DefenseClaw delivers security governance for agentic AI on OpenClaw, scanning skills, MCP servers, and plugins with Cisco AI Defense tools plus CodeGuard static analysis for secrets, injections, and vulns before execution. A Python CLI handles scans (`defenseclaw skill scan all`), block/allow lists, and alerts, while a Go gateway enforces runtime inspection on LLM prompts, tool calls, and network traffic via firewall rules. SIEM integration forwards events to Splunk or OTLP collectors.

Why is it gaining traction?

Agentic AI risks like malicious skills or prompt injections demand proactive controls, and DefenseClaw hooks into OpenClaw via TypeScript plugin for zero-config runtime blocking. It tackles github security scanning, github security policy, and security github actions natively, standing out from generic scanners by auto-quarantining high-severity findings and supporting permissive/strict modes. Devs grab it for seamless github copilot hygiene in AI repos.

Who should use this?

Security governance specialists and IT security and governance managers securing agentic AI deployments. Security governance officers focused on security governance risk and compliance, especially in security governance d365 or d365fo environments. Teams auditing security github repository and github security projects before production.

Verdict

Solid start for agentic AI defense with intuitive CLI and Splunk dashboards, but 1.0% credibility score and 167 stars signal early maturity—run e2e tests first. Ideal prototype for security governance model adopters; monitor github security advisories for updates.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.