Agent Warden

Name: Agent Warden
Author: shivdev

@shivdev/agent-warden · by shivdev

A firewall for AI agents. Detects and blocks prompt injection in email, web pages, documents and MCP tool results before they reach your model.

246/wk 5 tools GitHub npm PyPI

No known CVEs

MIT license

Actively maintained

Last commit 2d ago

Works with most clients

Transport: stdio, sse, http

5 tools · ~169 tok

Grade A · 0.1% of 200K ctx

MCPpedia last refreshed this data 3h ago

Edit this pageView history

Agent Warden is an MCP server that a firewall for AI agents. Detects and blocks prompt injection in email, web pages, documents and MCP tool results before they reach your model. Its tool list has not been published yet over stdio, sse and http, requires no API key, and scores 89/100 on MCPpedia's security, maintenance and efficiency rubric.

Browser AI / ML

Step 1

Install in your client

Config is the same across clients - only the file and path differ.

Supported in Claude Desktopstdio, sse, http · Node 18+

Paste into ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "warden": {
      "env": {
        "WARDEN_URL": "http://localhost:3000"
      },
      "args": [
        "@shivdev/agent-warden",
        "--upstream-cmd",
        "uvx mcp-server-fetch"
      ],
      "command": "bunx"
    }
  }
}

Are you the author?

Add this badge to your README to show your security score and help users find safe servers.

Embed in your READMEAbout badges →

[![MCPpedia Score](https://mcppedia.org/api/badge/agent-warden)](https://mcppedia.org/s/agent-warden)

Read me

What Agent Warden does

84.4% accuracy · 85.8% precision · 2,339 public samples · LLM judge off · 551 passing tests

Test This Server

Run this in your terminal to verify the server starts. Then let us know if it worked - your result helps other developers.

npx -y '@shivdev/agent-warden' 2>&1 | head -1 && echo "✓ Server started successfully"

After testing, let us know if it worked:

Scored, not listed

Why this score

Five weighted categories - click any category to see the underlying evidence.

Score breakdown

89/100across 5 weighted dimensions

How we score →

0255075100

−11

Security

Maintenance

Efficiency

Documentation

Compatibility

Categoriesclick a row to see evidence

Security

OSV.dev

No known CVEs.

Checked @shivdev/agent-warden against OSV.dev.

Inventory

Tools (5)

Click any tool to inspect its schema.

~169 tokens total

Community

Reviews

Be the first to review

Have you used this server?

Share your experience - it helps other developers decide.

How easy was setup?Did it work reliably?How was the documentation?

Frequently Asked Questions

Is Agent Warden safe to use?: Agent Warden has no known CVEs as of the latest MCPpedia security scan. Authentication requirements were not detected automatically - check the project's documentation before connecting in a shared environment. Licensed under MIT.
How do I install Agent Warden?: Agent Warden supports copy-paste install configs on its MCPpedia page for Claude Desktop, Cursor, and Claude Code. Scroll to the Quick Install section and select your client.
What can Agent Warden do?: Agent Warden provides 5 tools: Install, Prerequisites, Architecture, Benchmark, More. See the full tools list on the server page for descriptions and parameters.
What AI clients work with Agent Warden?: Agent Warden is compatible with claude-desktop, cursor, claude-code. It uses stdio and sse and http transport.
Is Agent Warden actively maintained?: Agent Warden is actively maintained - last commit was 2 days ago.

Similar servers

Others in browser / ai-ml

View all →

io.github.HUANGCHIHHUNGLeo/claude-real-video96

Let any LLM watch a video locally — and search everything it has ever watched.

2.2k 6

Modelcontextprotocol96

The official MCP server implementation for the Perplexity API Platform

2.6k 4

io.github.callstack/agent-device96

MCP server for mobile app automation: verify, control, and debug iOS, Android, TV, and desktop apps

5.1k 5

io.github.mims-harvard/tooluniverse96

2,700+ scientific tools for AI scientists: life science, research, literature, and more.

1.8k 4

MCP Security Weekly

Get CVE alerts and security updates for Agent Warden and similar servers.

Community

Discussion

Start a conversation

Ask a question, share a tip, or report an issue.

Has anyone used this with Cursor?How do you handle auth?Any alternatives?

{ "mcpServers": { "warden": { "env": { "WARDEN_URL": "http://localhost:3000" }, "args": [ "@shivdev/agent-warden", "--upstream-cmd", "uvx mcp-server-fetch" ], "command": "bunx" } } }

Frequently Asked Questions

Is Agent Warden safe to use?

Agent Warden has no known CVEs as of the latest MCPpedia security scan. Authentication requirements were not detected automatically - check the project's documentation before connecting in a shared environment. Licensed under MIT.

How do I install Agent Warden?

Agent Warden supports copy-paste install configs on its MCPpedia page for Claude Desktop, Cursor, and Claude Code. Scroll to the Quick Install section and select your client.

What can Agent Warden do?

Agent Warden provides 5 tools: Install, Prerequisites, Architecture, Benchmark, More. See the full tools list on the server page for descriptions and parameters.

What AI clients work with Agent Warden?

Agent Warden is compatible with claude-desktop, cursor, claude-code. It uses stdio and sse and http transport.

Is Agent Warden actively maintained?

Agent Warden is actively maintained - last commit was 2 days ago.