MCP (Model Context Protocol)
MCP (Model Context Protocol)
Shipbook supports MCP (Model Context Protocol) for programmatic access to logs and error data.
The MCP endpoint is available at: https://api.shipbook.io/mcp
Shipbook's MCP implements OAuth 2.1 for secure authentication — supporting both dynamic and static client registration. For MCP clients that don't support OAuth, you can authenticate with a token from the authentication keys section in the console. For more information on how to create and manage authentication keys, see Authentication Keys. See the Authentication section for full details.
Configuring MCP with AI Assistants
Claude
Shipbook's MCP uses OAuth, so you can connect securely without managing API tokens manually.
The connector uses OAuth, so your credentials are never shared with Claude.
Claude.ai (Connector)
This is the easiest way to connect — no tokens or configuration files needed.
- Go to claude.ai and open Settings
- Navigate to Connectors
- Click Add Connector and enter the URL:
https://api.shipbook.io/mcp - Click Connect — you'll be redirected to Shipbook to log in and authorize access
- Once authorized, Shipbook tools are available in your Claude conversations
Claude Code (CLI)
No tokens or configuration files needed — just add the server and log in.
claude mcp add --transport http shipbook-mcp https://api.shipbook.io/mcp
On first use, Claude Code will open your browser to log in and authorize access. Tokens are stored securely and refreshed automatically.
For more information view https://code.claude.com/docs/en/mcp#option-1:-add-a-remote-http-server
Cursor
One-Click Install
The easiest way to connect Shipbook MCP to Cursor — just click the button below:
This will automatically configure the Shipbook MCP server in Cursor. On first use, you'll be redirected to Shipbook to log in and authorize access via OAuth.
Manual Configuration
If you prefer to configure manually, open Cursor Settings (Cmd/Ctrl + ,), navigate to "Tools & MCP" → "Add MCP server", and add:
{
"mcpServers": {
"shipbook": {
"url": "https://api.shipbook.io/mcp",
"auth": {
"CLIENT_ID": "a1db8df5-ceb5-4013-8c35-af9e457c69b5"
}
}
}
}
This is saved at .cursor/mcp.json in your project directory.
Copilot (VS Code)
One-Click Install
This will automatically configure the Shipbook MCP server in VS Code. On first use, you'll be redirected to Shipbook to log in and authorize access.
Manual Configuration
Open the Command Palette (Cmd/Ctrl + Shift + P), run MCP: Add Server, choose "HTTP" and enter: https://api.shipbook.io/mcp
Or add directly to .vscode/mcp.json in your project:
{
"servers": {
"shipbook": {
"type": "http",
"url": "https://api.shipbook.io/mcp"
}
}
}
For more information view https://code.visualstudio.com/docs/copilot/customization/mcp-servers
Codex
Add the following to your ~/.codex/config.toml (or .codex/config.toml in your project root):
[mcp_servers.shipbook]
url = "https://api.shipbook.io/mcp"
Then log in:
codex mcp login shipbook
This will open your browser to log in and authorize access.
For more information view https://developers.openai.com/codex/mcp/
Google Antigravity
Google Antigravity supports custom MCP servers.
- Open the MCP store via the "..." dropdown at the top of the editor's agent panel
- Click on "Manage MCP Servers"
- Click on "View raw config"
- Add Shipbook to your
mcp_config.json:
{
"mcpServers": {
"shipbook": {
"url": "https://api.shipbook.io/mcp"
}
}
}
On first use, you'll be redirected to Shipbook to log in and authorize access.
For more information view https://antigravity.google/docs/mcp
Authentication
Shipbook's MCP endpoint supports multiple authentication methods to work with different MCP clients.
OAuth 2.1 (Recommended)
Shipbook implements OAuth 2.1 for secure, token-free authentication. Depending on your MCP client's capabilities, there are two modes:
-
Dynamic Client Registration — MCP clients that support OAuth 2.1 with dynamic client registration (e.g., Claude.ai Connectors) can automatically register and authenticate without any pre-configured client ID. The MCP client handles the entire OAuth flow on your behalf.
-
Static Client — For MCP clients that support OAuth but not dynamic registration (e.g., Cursor), use Shipbook's public client ID (
a1db8df5-ceb5-4013-8c35-af9e457c69b5). The client will redirect you to Shipbook to log in and authorize access.
In both cases, your credentials are never shared with the MCP client — authentication goes directly through Shipbook's OAuth flow.
API Key (Bearer Token)
For MCP clients that don't support OAuth at all (e.g., some CLI tools), you can authenticate using an API key passed as a Bearer token.
To get your API key, go to the Authentication Keys section in the Shipbook console. For more details, see Authentication Keys.
{
"mcpServers": {
"shipbook": {
"url": "https://api.shipbook.io/mcp",
"headers": {
"Authorization": "Bearer YOUR_AUTH_TOKEN"
}
}
}
}
Replace YOUR_AUTH_TOKEN with your authentication token from the console.
Available Tools
Once connected, the following tools are available to your AI assistant:
- get-account-apps — Lists all apps in your account with their name, appId, and platform.
- get-loglytics-errors — Retrieves errors from Loglytics for a specific app, with optional filtering by severity and app version.
- get-logs — Searches and retrieves logs for an app. Supports filtering by severity, user email, device, time range, session, and more.
Example Questions
Once configured, you can interact with the AI assistant to query your Shipbook logs and errors. The assistant will use the Shipbook MCP tools to access your data directly and can even help fix issues in your code when applicable.
Example queries:
- "Find logs for email [email protected] and fix any code issues if found"
- "Fix the issues found in Shipbook Loglytics"
The AI assistant leverages Shipbook MCP tools to query your logs and error data in real-time, providing insights into your application's behavior and issues. When applicable, it can also analyze the code and suggest fixes for problems identified in the logs.
To explore Shipbook's vision for the AI age of logging and how logs serve as a trusted interface for autonomous agents, check out the blog post Logs for AI Agents.