Start Here

New to all this? You are in the right place. In about two minutes you will know what SelfLabbs is, whether it is for you, and exactly how to use it — no jargon.

What is SelfLabbs, in one sentence?

SelfLabbs gives your AI agent live, real-world data — security, market, financial, weather, regulatory and blockchain information — so it stops guessing and starts checking.

Wait — what is an “AI agent”?

An AI agent is an AI assistant that can actually do things, not just chat: write code, research, automate tasks. You may already use one — Claude, Cursor, and Windsurf are all AI agents. The important part: agents can use tools. SelfLabbs makes tools.

And “MCP”?

MCP (Model Context Protocol) is simply the standard plug for adding new tools to your agent — like adding an app to a phone. Each SelfLabbs server is one of those tools your agent can plug into.

What SelfLabbs gives your agent

• Security Intel — check a package or dependency for known vulnerabilities before trusting it.
• Domain Intel — look up who owns a domain, its DNS, email setup, and tech stack.
• Market Intel — live foreign-exchange rates.
• Base Intel — read wallets, tokens, gas, and transactions on the Base blockchain.
• Package Intel — vet an npm, PyPI, or crates.io dependency: versions, popularity, health, and known advisories.
• Filings Intel — search SEC EDGAR: company filings, financials, and insider trades.
• Economy Intel — GDP, inflation, unemployment and more for any country (World Bank + US BLS).
• Weather & Geo Intel — US forecasts, weather alerts, earthquakes, elevation, and US address geocoding.
• Regulatory Intel — track US federal rules, notices, and executive orders from the Federal Register.

Is this for me?

If you build with AI agents — or want to — and you would rather they work with real, current data than guess, then yes. You do not need to be an expert. And if you do not have an agent yet, Step 1 below gets you one for free.

How to use it — 3 steps

Step 1 · Get an AI agent that supports MCP

Do not have one? The easiest free start is Claude Desktop or Cursor. Both support MCP out of the box.

Step 2 · Add a SelfLabbs server (one line, no API key)

Point your agent at a SelfLabbs server. For example, with Claude’s command line:

claude mcp add --transport http security-intel https://security.selflabbs.com/mcp

Or just paste that URL into your agent’s MCP / connector settings. That is it — your agent now has security tools.

Step 3 · Ask your agent to use it

Ask in plain English:

Check if the npm package left-pad has any known vulnerabilities.

Your agent calls SelfLabbs, gets live data, and answers. Done.

What does it cost?

See all the servers →

Part of SelfLabbs — keyless intelligence APIs for AI agents.