stevenkozeniesky02

stevenkozeniesky02 / agentsid-scanner

Public

Security scanner for MCP servers. Grades auth, permissions, injection risks, and tool safety. The Lighthouse of agent security.

agentsid ai-agents audit mcp scanner
13
0
100% credibility
Found Apr 03, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

AgentsID Scanner is a security tool that examines AI agent servers for vulnerabilities like prompt injections, missing permissions, and weak validation, delivering a graded report with recommendations.

How It Works

1
🛡️ Worry about your AI helper's safety

You build a server for AI agents and wonder if it's secure against tricks and risks.

2
📥 Grab the scanner easily

You pick up the scanner tool with a quick download, ready to check your server.

3
Choose your scan type
🏠
Local server

Point it at your running server on your computer.

🌐
Remote server

Enter the web address of your online server.

4
📊 Get your security report card

Watch as it checks everything and hands you a letter grade with details on issues.

5
💡 Spot the problems and tips

Read the clear list of risks and simple ways to make things safer.

✅ Server is now secure and strong

Fix the issues, recheck, and celebrate your top-grade safe AI helper.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 13 to 13 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 agentsid-scanner?

Agentsid-scanner is a JavaScript-based open source security scanner github tool for MCP servers, grading auth, permissions, injection risks, and tool safety with a simple letter grade A-F and detailed report. It scans tool descriptions for 11 prompt injection patterns, classifies tools by risk like destructive or credential access, and flags missing validation or secrets exposure—perfect for spotting vulnerabilities in AI agent toolsets. Run it via CLI like npx @agentsid/scanner -- your-server-command, or use the security github actions integration for automated checks.

Why is it gaining traction?

It stands out as a free security scanner software tailored for agent security, delivering instant reports on critical issues like unauthenticated shell tools or unbounded inputs, unlike generic github security scanning tools. Developers hook on the quick stdio/HTTP scans, JSON output for CI, and programmatic APIs, plus GitHub Action outputs for grades and findings that fail builds on poor scores. No setup hassle beats broader security scanner microsoft or servicenow options for MCP-specific risks.

Who should use this?

MCP server builders exposing tools to AI agents, like those integrating with github security copilot or custom agent projects. Teams in security github repositories running CI/CD via security github actions, or devs auditing remote MCP endpoints for injection and permissions before production. Ideal for agent devs checking github security advisories in tool definitions without spinning up full servers.

Verdict

Promising niche security scanner open source for MCP, but at 13 stars and 1.0% credibility score, it's early-stage with basic docs—test thoroughly before trusting in prod. Grab it if you're in agent security; skip for mature github security policy needs.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.