# protogrid

> Let your agent choose its own MCP servers, on evidence. protogrid searches every MCP server in the official registry by intent, probes every remote endpoint without credentials, scores trust and quality in the open, and hands your agent a configuration for 8 clients. Registry only: no proxying, no credential custody.

## Autonomy, with judgment

Your agent picks its own tools; you set the rules once. One search call carries them:

```
search("send an invoice", class=["R0", "R1"], min_quality=75, exclude_quality_flags=["known-vulns"], owner_verified=True)
```

Signals an agent reads for every server: answers right now, needs a key or a sign-in (connection class R0, R1, R2, L0), quality 0-100, trust 0-100, tool-definition changes, known advisories in dependencies, owner verified, 2026-07-28 protocol.

## Connect protogrid to your agent

### Claude Code

terminal:

```bash
claude mcp add --transport http -s user protogrid https://api.protogrid.dev/mcp
```

### Codex

~/.codex/config.toml:

```toml
[mcp_servers.protogrid]
url = "https://api.protogrid.dev/mcp"
```

### OpenCode

~/.config/opencode/opencode.json:

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "protogrid": {
      "type": "remote",
      "url": "https://api.protogrid.dev/mcp",
      "enabled": true
    }
  }
}
```

### Cursor

~/.cursor/mcp.json:

```json
{
  "mcpServers": {
    "protogrid": {
      "url": "https://api.protogrid.dev/mcp"
    }
  }
}
```

### VS Code

.vscode/mcp.json:

```json
{
  "servers": {
    "protogrid": {
      "type": "http",
      "url": "https://api.protogrid.dev/mcp"
    }
  }
}
```

### Other clients

mcpServers JSON:

```json
{
  "mcpServers": {
    "protogrid": {
      "type": "http",
      "url": "https://api.protogrid.dev/mcp"
    }
  }
}
```

## Use it as an agent

- REST: `https://api.protogrid.dev/v1/search?q=<intent>&class=R0,R1`, then `/v1/servers/{name}/connection?target=mcpServers`
- MCP: `https://api.protogrid.dev/mcp` (streamable HTTP, no auth; tools search, get_server, list_tools, get_connection)
- Docs: https://docs.protogrid.dev/llms.txt

## For MCP server owners

Claim your servers (GitHub sign-in or DNS) for owner verified, hourly probes (3 servers on the free plan), email alerts, signed webhooks and the README badge: https://protogrid.dev/owners

## Questions

### I publish an MCP server. What do I get?

Claim it with GitHub or DNS and get owner verified, hourly probes, email alerts, signed webhooks and the README badge, free with 3 servers monitored hourly. [Claim your servers](https://protogrid.dev/auth/login?next=/account/servers).

### Can my agent choose servers without asking me?

Yes, within the rules you set: connection class, minimum quality, no known advisories, a verified owner. Servers that need a key use the ones you already gave it; OAuth asks you once.

### Is protogrid free?

Yes. Search, descriptors, configurations, the MCP server and checks need no account. A free account adds an API key with higher limits.

### Do you see my keys or tokens?

Never. protogrid is a registry, not a gateway: secrets stay placeholders and your agent connects directly.

### How are scores calculated?

From signals anyone can observe, with published weights. No audit is implied; every check shows its evidence.

### How is this different from the official MCP registry?

We mirror it every hour, then add what an agent needs to choose and connect: live probes, search by intent, scores and configurations.

## Pages

- /search.md?q=<intent>: results as Markdown
- /servers/<name>.md: a server page as Markdown
- /owners.md: what claiming a server gives its owner