Docs

Overview

Badlands is a secure gateway for AI agent skills. It gives your AI tools — Claude Code, Cursor, Cline — safe, controlled access to your business systems, without credentials ever leaving your infrastructure.

Instead of hardcoding API keys into your local IDE or giving AI direct database access, Badlands acts as a managed proxy. You deploy skills (Python or TypeScript functions) to Badlands, and your AI clients call them through a secure endpoint with a token — nothing sensitive touches the client.

Badlands skills are production-ready integrations maintained by us. You can also deploy your own custom functions.

Docs

Quick Start

Get a skill running in under 5 minutes.

1

Contact us

Email info@badlandslabs.com to get your account and Badlands token set up.

2

Pick a skill

Choose from our library — ERP query, CRM sync, SQL database, Slack notifications, and more.

3

Connect via MCP or API

Drop in the MCP config (for Claude Code / Cursor) or call the REST endpoint directly. Your token is the only credential needed.

4

Ask your AI

Tell Claude Code or Cursor to use the skill. Badlands runs it server-side and returns the result — your credentials never leave our infrastructure.

Docs

Skill Library

Production-ready skills maintained by the Badlands team. No integration boilerplate required.

ERP Integrations

SAP, Oracle, NetSuite — query and write data without raw SQL credentials on the client.

CRM Sync

Salesforce, HubSpot — read and update records, trigger workflows, pull reports.

Real-time Data

SQL databases, REST APIs, webhooks — expose live data without exposing your schema.

Team Notifications

Slack, Teams, email — AI can send updates and alerts without having your credentials.

Don't see what you need? Contact us — we can build and maintain a custom skill for your stack.

Docs

Authentication

Badlands uses token-based authentication. Your AI client only needs a single token — no API keys, no database passwords.

Via MCP

Set the BADLANDS_TOKEN environment variable in your MCP config:

{
  "mcpServers": {
    "badlands": {
      "command": "npx",
      "args": ["@badlandslabs/mcp"],
      "env": {
        "BADLANDS_TOKEN": "sk_••••••••"
      }
    }
  }
}

Via REST API

Include your token in the request payload:

{
  "endpoint": "https://api.badlandslabs.com/v1/execute",
  "payload": {
    "id": "your_tenant",
    "auth": "sk_••••••••",
    "skill": "skill_name",
    "request": { ... }
  }
}

Never share your token. Badlands will never ask for your database password or internal API keys.

MCP

MCP Overview

Badlands natively supports the Model Context Protocol (MCP), making it compatible with Claude Code, Cursor, and any MCP-compatible client.

MCP lets you expose Badlands skills as tools that your AI can discover and call directly — without any custom prompt engineering. The skill does the work, Badlands handles security, routing, and credential isolation.

MCP

Setup

Install the Badlands MCP server and configure it with your token.

1. Install the package

npx @badlandslabs/mcp

2. Add to your MCP config

Find your editor's MCP config file and add:

{
  "mcpServers": {
    "badlands": {
      "command": "npx",
      "args": ["@badlandslabs/mcp"],
      "env": {
        "BADLANDS_TOKEN": "sk_••••••••"
      }
    }
  }
}

3. Restart your editor

Once configured, Badlands skills appear as available tools in Claude Code or Cursor. No additional setup needed.

MCP

Configuration Reference

All available MCP configuration options.

Setting	Description	Required
BADLANDS_TOKEN	Your Badlands authentication token	Yes
BADLANDS_ENDPOINT	Custom API endpoint (defaults to api.badlandslabs.com)	No

Legal

Privacy Policy

Last updated: June 2026

Badlands Labs ("we", "our", or "us") is committed to protecting your privacy. This Privacy Policy explains how we collect, use, and safeguard information when you use our service.

Information We Collect

We collect minimal information necessary to provide the service: your email address (for account management), your Badlands token, and anonymized usage metrics. We do not collect the content of your AI requests or the data returned by your skills.

Data Storage

Your credentials and skill configurations are stored securely in our infrastructure. We do not sell, share, or transfer your personal data to third parties. All data is hosted on GCP in Canadian data centers.

PIPEDA Compliance

As a Canadian company, we comply with the Personal Information Protection and Electronic Documents Act (PIPEDA). You have the right to access, correct, and request deletion of your personal information. Contact us at info@badlandslabs.com.

Contact

For privacy-related questions: info@badlandslabs.com

Legal

Terms of Service

Last updated: June 2026

By using Badlands Labs, you agree to these terms. If you do not agree, do not use the service.

Service Description

Badlands Labs provides a secure gateway service that routes AI tool calls between client applications and backend systems. We do not own, access, or store the data flowing through our infrastructure beyond what is necessary for routing and logging.

Acceptable Use

You may not use Badlands to: transmit malware, illegal content, or material that violates third-party rights; attempt to circumvent security controls; or resell the service without authorization.

Disclaimers

OUR SERVICES ARE PROVIDED "AS IS" AND "AS AVAILABLE" WITHOUT WARRANTIES OF ANY KIND. WE DO NOT WARRANT THAT THE SERVICE WILL BE UNINTERRUPTED OR ERROR-FREE.

Limitation of Liability

BADLANDS LABS SHALL NOT BE LIABLE FOR ANY INDIRECT, INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES ARISING FROM YOUR USE OF THE SERVICE.

Contact

For questions: info@badlandslabs.com