We Built a Trust API. ChatGPT Found It on Its Own and Makes 1,400+ Checks Daily.

Something unexpected happened at Nerq.

We built a trust verification API for AI agents. We published an `llms.txt` file. We made the API free and keyless. Then we watched the logs.

Within days, ChatGPT started calling our API. Not because we asked OpenAI. Not because we had a partnership. ChatGPT discovered our API through `llms.txt` and started making trust verification calls autonomously.

The Numbers

- **1,400+** daily preflight checks from AI systems
- **ChatGPT** is the single largest consumer
- **Perplexity**, **Claude**, and other AI systems also discovered us
- **Zero** of these integrations were arranged by humans

How It Happened

Step 1: We Published llms.txt

Following the emerging `llms.txt` standard, we published a machine-readable description of our API at `nerq.ai/llms.txt`. It includes:

- What Nerq does (trust verification for AI agents)
- API endpoints and how to use them
- A decision tree for AI systems ("Is this agent safe?")

Step 2: AI Systems Found It

AI systems that browse the web discovered `llms.txt`. When a user asked "Is langchain safe to use?" or "What's the trust score for crewai?", the AI system found Nerq's API and called it.

Step 3: They Kept Coming Back

The key insight: once an AI system learns about a useful API, it remembers. ChatGPT didn't just call us once — it integrated Nerq into its decision-making for agent-related queries. Every day, 1,400+ trust checks.

What This Means

The Machine Economy Is Real

AI systems are now discovering, evaluating, and integrating with other services autonomously. This isn't science fiction — it's happening in our server logs right now.

llms.txt Is the New robots.txt

Just as `robots.txt` told search engines how to crawl your site, `llms.txt` tells AI systems what your service does and how to use it. If you're building an API, you need one.

Trust Is Infrastructure

When ChatGPT checks Nerq before recommending an AI agent, it's using trust verification as infrastructure. The same way a browser checks SSL certificates before connecting, AI systems are starting to check trust scores before recommending tools.

Machine-to-Machine Is the Growth Channel

Our fastest-growing channel isn't Google Search or Product Hunt. It's machine-to-machine discovery. AI systems recommending us to other AI systems.

The Stack That Made This Possible

1. **llms.txt** — machine-readable API description
2. **Free, keyless API** — zero friction for machine consumers
3. **Fast responses** — sub-100ms for preflight checks
4. **Structured output** — JSON responses that AI systems can parse
5. **MCP server** — native integration for the MCP ecosystem

Try It Yourself

Ask ChatGPT: "What's the trust score for langchain on Nerq?"

Or call the API directly:
```bash
curl https://nerq.ai/v1/preflight?target=langchain
```

Or add the MCP server to your AI system for native trust verification.

What's Next

We're seeing the early signs of an autonomous trust layer forming. AI systems checking other AI systems before interacting. Trust scores as machine-readable infrastructure. Autonomous discovery through llms.txt and MCP.

Nerq is positioning to be the trust oracle for this machine economy. Independent, quantitative, and machine-first.

---

*[Nerq](https://nerq.ai) indexes 204,000+ AI agents with independent trust scores. Free API, MCP server, GitHub Action, and CLI.*