Skip to main content
Glama
starskrime

MCP Action Firewall

by starskrime

πŸ”₯ MCP Action Firewall

Python 3.12+ License: MIT MCP Compatible

Works with any MCP-compatible agent

Claude Cursor Windsurf OpenAI Gemini OpenClaw

A transparent MCP proxy that intercepts dangerous tool calls and requires OTP-based human approval before execution. Acts as a circuit breaker between your AI agent and any MCP server.

How It Works

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”    stdin/stdout    β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”    stdin/stdout    β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚ AI Agent β”‚ ◄────────────────► β”‚   MCP Action     β”‚ ◄────────────────► β”‚ Target MCP Serverβ”‚
β”‚ (Claude) β”‚                    β”‚   Firewall       β”‚                    β”‚ (e.g. Stripe)    β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜                    β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜                    β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                                        β”‚
                                   Policy Engine
                                  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
                                  β”‚ Allow? Block? β”‚
                                  β”‚ Generate OTP  β”‚
                                  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

MCP servers don't run like web servers β€” there's no background process on a port. Instead, your AI agent (Claude, Cursor, etc.) spawns the MCP server as a subprocess and talks to it over stdin/stdout. When the chat ends, the process dies.

The firewall inserts itself into that chain:

Without firewall:
  Claude ──spawns──► mcp-server-stripe

With firewall:
  Claude ──spawns──► mcp-action-firewall ──spawns──► mcp-server-stripe

So you just replace the server command in your MCP client config with the firewall, and tell the firewall what the original command was:

Before (direct):

{ "command": "uvx", "args": ["mcp-server-stripe", "--api-key", "sk_test_..."] }

After (wrapped with firewall):

{ "command": "uv", "args": ["run", "mcp-action-firewall", "--target", "mcp-server-stripe --api-key sk_test_..."] }

Then the firewall applies your security policy:

  1. βœ… Safe calls (e.g. get_balance) β†’ forwarded immediately

  2. πŸ›‘ Dangerous calls (e.g. delete_user) β†’ blocked, OTP generated

  3. πŸ”‘ Agent asks user for the code β†’ user replies β†’ agent calls firewall_confirm β†’ original action executes

Related MCP server: sovr-mcp-proxy

Installation

pip install mcp-action-firewall
# or
uvx mcp-action-firewall --help

Quick Start β€” MCP Client Configuration

Add the firewall as a wrapper around any MCP server in your client config:

{
  "mcpServers": {
    "stripe": {
      "command": "uv",
      "args": ["run", "mcp-action-firewall", "--target", "mcp-server-stripe --api-key sk_test_abc123"]
    }
  }
}

That's it. Everything after --target is the full shell command to launch the real MCP server β€” including its own flags like --api-key. The firewall doesn't touch those args, it just spawns the target and sits in front of it.

More Examples

{
  "mcpServers": {
    "stripe": {
      "command": "uv",
      "args": [
        "run", "mcp-action-firewall",
        "--target", "uvx mcp-server-stripe --api-key sk_test_...",
        "--name", "stripe"
      ]
    },
    "database": {
      "command": "uv",
      "args": [
        "run", "mcp-action-firewall",
        "--target", "uvx mcp-server-postgres --connection-string postgresql://...",
        "--name", "database",
        "--config", "/path/to/my/firewall_config.json"
      ]
    }
  }
}
{
  "mcpServers": {
    "github": {
      "command": "uvx",
      "args": [
        "mcp-action-firewall",
        "--target", "npx @modelcontextprotocol/server-github"
      ]
    }
  }
}

The OTP Flow

When the agent tries to call a blocked tool, the firewall returns a structured response:

{
  "status": "PAUSED_FOR_APPROVAL",
  "message": "⚠️ The action 'delete_user' is HIGH RISK and has been locked by the Action Firewall.",
  "action": {
    "tool": "delete_user",
    "arguments": { "id": 42 }
  },
  "instruction": "To unlock this action, you MUST ask the user for authorization.\n\n1. Show the user the following and ask for approval:\n   Tool: **delete_user**\n   Arguments:\n{\"id\": 42}\n\n2. Tell the user: 'Please reply with approval code: **9942**' to allow this action, or say no to cancel.\n3. STOP and wait for their reply.\n4. When they reply with '9942', call the 'firewall_confirm' tool with that code.\n5. If they say no or give a different code, do NOT retry."
}

Argument visibility guarantee: The arguments shown to the user are frozen at interception time β€” they are taken from the original blocked call, not from what the agent passes to firewall_confirm. The agent cannot change the arguments after the OTP is issued.

The firewall_confirm tool is automatically injected into the server's tool list:

{
  "name": "firewall_confirm",
  "description": "Call this tool ONLY when the user provides the correct 4-digit approval code to confirm a paused action.",
  "inputSchema": {
    "type": "object",
    "properties": {
      "otp": {
        "type": "string",
        "description": "The 4-digit code provided by the user."
      }
    },
    "required": ["otp"]
  }
}

Configuration

The firewall ships with sensible defaults. Override with --config:

{
  "global": {
    "allow_prefixes": ["get_", "list_", "read_", "fetch_"],
    "block_keywords": ["delete", "update", "create", "pay", "send", "transfer", "drop", "remove", "refund"],
    "default_action": "block",
    "otp_attempt_count": 1
  },
  "servers": {
    "stripe": {
      "allow_prefixes": [],
      "block_keywords": ["refund", "charge"],
      "default_action": "block"
    },
    "database": {
      "allow_prefixes": ["select_"],
      "block_keywords": ["drop", "truncate", "alter"],
      "default_action": "block"
    }
  }
}

Rule evaluation order:

  1. Tool name starts with an allow prefix β†’ ALLOW

  2. Tool name contains a block keyword β†’ BLOCK (OTP required)

  3. No match β†’ fallback to default_action

otp_attempt_count β€” maximum number of failed OTP attempts before the pending action is permanently locked out. Defaults to 1 (any wrong code cancels the request). Increase for more forgiving UX, keep at 1 for maximum security.

Per-server rules extend (not replace) the global rules. Use --name stripe to activate server-specific overrides.

CLI Reference

--target (required)

The full command to launch the real MCP server. This is the server you want to protect:

mcp-action-firewall --target "mcp-server-stripe --api-key sk_test_abc123"
mcp-action-firewall --target "npx @modelcontextprotocol/server-github"
mcp-action-firewall --target "uvx mcp-server-postgres --connection-string postgresql://localhost/mydb"

--name (optional)

Activates per-server rules from your config. Without it, only global rules apply:

mcp-action-firewall --target "mcp-server-stripe" --name stripe

--config (optional)

Custom config file path. Without it, uses firewall_config.json in your current directory, or the bundled defaults:

mcp-action-firewall --target "mcp-server-stripe" --config /path/to/my_rules.json

-v / --verbose (optional)

Turns on debug logging (written to stderr, won't interfere with MCP traffic):

mcp-action-firewall --target "mcp-server-stripe" -v

Project Structure

src/mcp_action_firewall/
β”œβ”€β”€ __init__.py          # Package version
β”œβ”€β”€ __main__.py          # python -m support
β”œβ”€β”€ server.py            # CLI entry point
β”œβ”€β”€ proxy.py             # JSON-RPC stdio proxy
β”œβ”€β”€ policy.py            # Allow/block rule engine
β”œβ”€β”€ state.py             # OTP store with TTL
└── default_config.json  # Bundled default rules

Try It β€” Interactive Demo

See the firewall in action without any setup:

git clone https://github.com/starskrime/mcp-action-firewall.git
cd mcp-action-firewall
uv sync
uv run python demo.py

The demo simulates an AI agent and walks you through the full OTP flow:

  1. βœ… Safe call (get_balance) β†’ passes through instantly

  2. πŸ›‘ Dangerous call (delete_user) β†’ blocked, OTP generated

  3. πŸ”‘ You enter the code β†’ action executes after approval

Known Limitations

Argument Inspection

The firewall matches on tool names only, not argument values. This means a tool like get_data({"sql": "DROP TABLE users"}) would pass if get_ is in your allow list, because the policy engine only sees get_data.

Workaround: Use explicit tool names in your allow/block lists and set "default_action": "block" so unrecognized tools require approval.

🚧 Roadmap: Argument-level inspection (scanning argument values against block_keywords) is planned for a future release.

Development

# Install dev dependencies
uv sync

# Run tests
uv run pytest tests/ -v

# Run the firewall locally
uv run mcp-action-firewall --target "your-server-command" -v

License

MIT

Install Server
A
license - permissive license
C
quality
C
maintenance

Maintenance

–Maintainers
–Response time
0dRelease cycle
2Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • F
    license
    -
    quality
    -
    maintenance
    A transparent proxy and execution firewall that intercepts and audits AI agent tool calls against configurable security policies before forwarding them to downstream MCP servers. It provides safe execution environments with features like data redaction, anti-loop protection, and unified alert dispatching.
    Last updated
  • A
    license
    -
    quality
    D
    maintenance
    Human-in-the-Loop authorization gateway for AI Agents. Securely pause MCP workflows and route high-risk actions to human approvers via Slack or Email.
    Last updated
    51
    1
    MIT
  • A
    license
    A
    quality
    C
    maintenance
    Security-enforcing MCP proxy that sits between an AI agent and any number of downstream MCP servers, intercepting every tool call through a capability-token policy gateway that can allow, deny, or escalate to human approval before the call reaches any real tool. It also exposes built-in operator tools for approval workflows, audit trail queries, token management, voice/HUD output, and hierarchical
    Last updated
    21
    11
    Apache 2.0

View all related MCP servers

Related MCP Connectors

  • Security firewall for AI agents β€” scans MCP calls for injection, secrets, and risks.

  • Runtime permission, approval, and audit layer for AI agent tool execution.

  • Crypto transaction firewall and risk tools for MCP agents.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/starskrime/mcp-action-firewall'

If you have feedback or need assistance with the MCP directory API, please join our Discord server