> ## Documentation Index
> Fetch the complete documentation index at: https://docs.limitguard.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Discovery

> Machine-readable discovery for AI agents — MCP, A2A, x402, and llms.txt

LimitGuard publishes machine-readable discovery files for three AI agent protocols. All discovery endpoints are **free and require no authentication**.

## Discovery Endpoints

| Endpoint                        | Protocol                     | Purpose                                     |
| ------------------------------- | ---------------------------- | ------------------------------------------- |
| `GET /.well-known/x402.json`    | x402 Bazaar                  | AI marketplace service listing with pricing |
| `GET /.well-known/agent.json`   | A2A (Agent-to-Agent)         | Agent Card for capability discovery         |
| `GET /.well-known/mcp.json`     | MCP (Model Context Protocol) | Tool manifest for LLM agents                |
| `GET /.well-known/security.txt` | RFC 9116                     | Security contact information                |
| `GET /llms.txt`                 | llms.txt                     | Concise LLM-readable API summary            |
| `GET /llms-full.txt`            | llms.txt                     | Full LLM-readable API reference             |

<Note>
  All `/.well-known/*` requests bypass x402 payment middleware completely — no payment or API key needed.
</Note>

## x402 Bazaar Listing

Enables AI agents to discover available endpoints and their payment requirements automatically.

```bash theme={null}
curl https://api.limitguard.ai/.well-known/x402.json
```

```json Response theme={null}
{
  "name": "LimitGuard Trust Intelligence API",
  "description": "Entity verification and risk scoring for EU businesses",
  "endpoints": [
    {
      "path": "/v1/entity/check",
      "method": "POST",
      "price_usdc": "0.05",
      "chain": "eip155:8453",
      "currency": "USDC"
    }
  ],
  "payment_methods": [
    {"type": "x402", "version": "2"},
    {"type": "api_key"}
  ]
}
```

Also available at `/v1/sales/listing` for the same data.

## A2A Agent Card

Google's Agent-to-Agent (A2A) protocol for agent capability discovery.

```bash theme={null}
curl https://api.limitguard.ai/.well-known/agent.json
```

Returns capabilities, skills, and how to invoke LimitGuard as an A2A agent. Other agents can discover what LimitGuard can do and how to interact with it programmatically.

## MCP Manifest

Model Context Protocol tools for LLM agents (Claude, GPT-4, etc.).

```bash theme={null}
curl https://api.limitguard.ai/.well-known/mcp.json
```

Returns tool definitions that LLM agents can use to call LimitGuard endpoints as native tools. See the [AI Agent Integration guide](/guides/ai-agent-integration) for usage examples.

## llms.txt

Standard machine-readable format for LLM consumption.

```bash theme={null}
# Concise summary for quick context loading
curl https://api.limitguard.ai/llms.txt

# Full reference with all endpoints, pricing, and examples
curl https://api.limitguard.ai/llms-full.txt
```

## Cross-Sell Recommendations

AI agents calling one endpoint can discover complementary services:

```bash theme={null}
curl "https://api.limitguard.ai/v1/sales/recommendations?current_endpoint=/v1/risk/score&trust_score=45"
```

Returns suggestions like "low risk score -> consider full entity check for more detail".

## Integration Example

An AI agent discovering and using LimitGuard autonomously:

<CodeGroup>
  ```python Python theme={null}
  import httpx

  # Step 1: Discover capabilities
  listing = httpx.get("https://api.limitguard.ai/.well-known/x402.json").json()
  entity_check = next(e for e in listing["endpoints"] if e["path"] == "/v1/entity/check")

  # Step 2: Check price
  price_usdc = float(entity_check["price_usdc"])  # 0.05

  # Step 3: Build x402 payment and call
  from my_wallet import build_x402_payment
  x_payment = build_x402_payment(amount=int(price_usdc * 1_000_000))

  response = httpx.post(
      "https://api.limitguard.ai/v1/entity/check",
      headers={"X-PAYMENT": x_payment},
      json={"entity_name": "Acme Corp BV", "country": "NL"},
  )
  print(response.json()["trust_score"])  # 87
  ```

  ```javascript JavaScript theme={null}
  // Step 1: Discover capabilities
  const listing = await fetch("https://api.limitguard.ai/.well-known/x402.json")
    .then((r) => r.json());
  const entityCheck = listing.endpoints.find((e) => e.path === "/v1/entity/check");

  // Step 2: Check price
  const priceUsdc = parseFloat(entityCheck.price_usdc); // 0.05

  // Step 3: Build x402 payment and call
  const xPayment = await buildX402Payment(priceUsdc * 1_000_000);

  const response = await fetch("https://api.limitguard.ai/v1/entity/check", {
    method: "POST",
    headers: { "X-PAYMENT": xPayment, "Content-Type": "application/json" },
    body: JSON.stringify({ entity_name: "Acme Corp BV", country: "NL" }),
  });
  const result = await response.json();
  console.log(result.trust_score); // 87
  ```
</CodeGroup>