MCP Server

HookDeploy exposes a remote Model Context Protocol (MCP) server, letting you manage your webhook infrastructure directly from AI assistants — create endpoints, inspect captured requests, replay webhooks, and check usage without leaving your chat interface.

Available on Team and Enterprise plans.

Server URL

https://mcp.hookdeploy.dev/mcp

Authentication

Use your HookDeploy API key (create one in Settings → API Keys). The key is passed as a Bearer token.

Available tools

Connect from Claude.ai

Requires a Claude Team or Enterprise plan. (Pro/Max users: use Settings → Connectors → Add custom connector.)

For Team plans — only a workspace Owner can add connectors. Members connect individually once the connector is added.

1. Go to Settings → Connectors → Add custom connector (Pro/Max) or Organization Settings → Connectors → Add (Team/Enterprise Owner).
2. In the URL field, enter:

   https://mcp.hookdeploy.dev/mcp

3. In the Authorization field (if shown), enter:

   Bearer hd_live_xxxx

   Replace hd_live_xxxx with your actual API key.
4. Name it HookDeploy and click Add.
5. Click Connect to activate.

Once connected, Claude can use HookDeploy tools in any conversation. Try:

“List my HookDeploy endpoints” “Create a new endpoint called Stripe production” “Show me the last 10 requests on my Stripe endpoint”

Connect from Claude Desktop

Claude Desktop connects to remote MCP servers via the mcp-remote proxy. Node.js 18+ is required.

1. Open your Claude Desktop config file:

   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Windows: %APPDATA%\Claude\claude_desktop_config.json

2. Add the following entry (create the file if it doesn’t exist):

{
  "mcpServers": {
    "hookdeploy": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.hookdeploy.dev/mcp?api_key=hd_live_xxxx"
      ]
    }
  }
}

Replace hd_live_xxxx with your actual API key.

3. Save the file and restart Claude Desktop.

Claude Desktop will load HookDeploy as an available tool set. You’ll see a hammer icon in the chat interface when tools are active.

Connect from ChatGPT

Requires ChatGPT Plus, Pro, or a Business/Enterprise workspace with Developer Mode enabled.

1. In ChatGPT, go to Settings → Apps & Connectors → Advanced settings and toggle Developer Mode on.
2. Go to Settings → Apps & Connectors → Add new connector.
3. Enter:
   • Name: HookDeploy
   • MCP Server URL: https://mcp.hookdeploy.dev/mcp
   • Authentication: Select API Key or Bearer Token, and enter your hd_live_xxxx key.
4. Click Save.

The connector is active for the current conversation. To use it in a new chat, enable it from the connector list.

Note: ChatGPT renamed “connectors” to “apps” in December 2025. The setting is now under Apps & Connectors.

Connect from Perplexity

Requires Perplexity Pro or Max plan.

1. Open Perplexity and go to Settings → Connectors → Add Connector.
2. Choose Remote MCP (or the Simple / URL tab depending on your client version).
3. Enter:
   • Server Name: HookDeploy
   • URL: https://mcp.hookdeploy.dev/mcp
   • API Key / Auth: your hd_live_xxxx key
4. Click Save and confirm the connector shows Running status.
5. In a new chat, toggle HookDeploy on under Sources or the tools panel.

Connect from OpenClaw

OpenClaw is an open-source self-hosted AI assistant that supports MCP via HTTP.

Add HookDeploy to your OpenClaw config (~/.openclaw/config.yaml or your config path):

mcp:
  servers:
    hookdeploy:
      transport: http
      url: "https://mcp.hookdeploy.dev/mcp"
      headers:
        Authorization: "Bearer hd_live_xxxx"

Replace hd_live_xxxx with your API key. Restart OpenClaw to load the server.

You can also pass the key in the URL if your OpenClaw version doesn’t support custom headers:

url: "https://mcp.hookdeploy.dev/mcp?api_key=hd_live_xxxx"

Connect via the Anthropic API

If you’re building an agent using the Anthropic API, you can attach the HookDeploy MCP server directly:

import anthropic

client = anthropic.Anthropic()

response = client.beta.messages.create(
    model="claude-opus-4-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "List my HookDeploy endpoints"}],
    mcp_servers=[
        {
            "type": "url",
            "url": "https://mcp.hookdeploy.dev/mcp",
            "name": "hookdeploy",
            "authorization_token": "hd_live_xxxx",
        }
    ],
    betas=["mcp-client-2025-11-20"],
)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic();
const response = await client.beta.messages.create({
  model: "claude-opus-4-5",
  max_tokens: 1024,
  messages: [{ role: "user", content: "List my HookDeploy endpoints" }],
  mcp_servers: [
    {
      type: "url",
      url: "https://mcp.hookdeploy.dev/mcp",
      name: "hookdeploy",
      authorization_token: "hd_live_xxxx",
    },
  ],
  betas: ["mcp-client-2025-11-20"],
});

Security

Your API key grants full access to your HookDeploy organization. Use a dedicated key for AI assistants so you can revoke it independently. Create keys with the minimum permissions needed — a viewer or developer role key covers all read operations; only admin/super_admin keys can manage members.

Never paste your API key directly into a chat message — always configure it in the connector auth settings.

Troubleshooting

“Missing HookDeploy API key” — The MCP client isn’t passing your API key. Double-check the Authorization header or api_key query param in your connector config.

“HookDeploy API error: plan_limit” — You’ve hit a plan limit. Check your usage in the dashboard.

“HookDeploy API error: forbidden” — Your API key doesn’t have permission for this action. Member management requires an admin-level key.

Connector connects but no tools appear — Confirm you’re on a Team or Enterprise plan. If you just upgraded, try disconnecting and reconnecting the connector.