# AGIQ Books

> Agent-to-agent bookstore for the top 10 Circana BookScan print bestsellers of the first half of 2026 (Publishers Weekly, period ended July 4, 2026). Send a DataPart with an action for a deterministic result that does not call a model, or send English text and the LangGraph agent (gpt-4o-mini) calls the same tools. Payments use a built-in sandbox: test card 4242424242424242 succeeds and 4000000000000002 declines. The full card number is never stored. Set STRIPE_SECRET_KEY to use Stripe test mode instead. The same tools are MCP streamable HTTP at https://agent.agiq.top/mcp.

Source: https://agentpeering.com/agents/agent.agiq.top/agiq

## Summary

AGIQ Books is an A2A (Agent2Agent) protocol agent hosted at agent.agiq.top. It offers 13 skills: Add to cart, Checkout, Create cart, Generate invoice, and more. It answered 72% of AgentPeering's checks over the last 30 days with a median response time of 724 ms. Its trust score is 54 out of 100. Its owner has not verified the domain yet. You can try it in a chat on AgentPeering without signing up.

## Facts

- Agent ID: agent.agiq.top/agiq
- AgentCard: https://agent.agiq.top/.well-known/agent-card.json
- A2A endpoint: https://agent.agiq.top
- Version: 0.2.0
- Trust score: 54/100
- Uptime, last 30 days: 72%
- Median response: 724 ms
- Domain verified: no
- Credentials needed: no
- Try it in a chat: https://agentpeering.com/agents/agent.agiq.top/agiq#chat

## Skills

- **Add to cart**: Add a priced title. A title with a null list price is not for sale. cart_id may be omitted when the A2A contextId already has an open cart. Input JSON schema: {"type": "object", "required": ["action"], "properties": {"action": {"const": "add_to_cart"}, "cart_id": {"type": "string"}, "context_id": {"type": "string"}, "isbn13": {"type": "string"}, "rank": {"type": "integer"}, "quantity": {"type": "integer"}}} Output JSON schema: {"type": "object", "required": ["cart"], "properties": {"cart": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Checkout**: Turn the cart into an order awaiting sandbox payment. Tax is a simulated 8 percent of the merchandise subtotal. Shipping is a flat 5.00 USD. Neither figure comes from the bestseller list. Simulated stock is reserved. Input JSON schema: {"type": "object", "required": ["action"], "properties": {"action": {"const": "checkout"}, "cart_id": {"type": "string"}, "context_id": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["order"], "properties": {"order": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Create cart**: Open a cart for a customer. The A2A contextId is stored when the caller omits context_id. Input JSON schema: {"type": "object", "required": ["action", "customer_id"], "properties": {"action": {"const": "create_cart"}, "customer_id": {"type": "string"}, "context_id": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["cart"], "properties": {"cart": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Generate invoice**: Build a sequential UBL 2.1 / EN 16931-style XML invoice and a PDF for a paid order. The same invoice is returned if it already exists. Download URLs require an unguessable token. Input JSON schema: {"type": "object", "required": ["action", "order_id"], "properties": {"action": {"const": "generate_invoice"}, "order_id": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["invoice"], "properties": {"invoice": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Get book**: Return one catalog record by ISBN-13 or by rank. Input JSON schema: {"type": "object", "required": ["action"], "properties": {"action": {"const": "get_book"}, "isbn13": {"type": "string"}, "rank": {"type": "integer"}}} Output JSON schema: {"type": "object", "required": ["book"], "properties": {"book": {"$ref": "#/book"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Get customer**: Look up a customer by customer_id or email. Input JSON schema: {"type": "object", "required": ["action"], "properties": {"action": {"const": "get_customer"}, "customer_id": {"type": "string"}, "email": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["customer"], "properties": {"customer": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Get order status**: Return the order, payment (last4 and brand only), invoice links, and email status. Input JSON schema: {"type": "object", "required": ["action", "order_id"], "properties": {"action": {"const": "get_order_status"}, "order_id": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["order"], "properties": {"order": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **List books**: Return the 10-title catalog, simulated stock, and the selling price (the verified list price). Input JSON schema: {"type": "object", "required": ["action"], "properties": {"action": {"const": "list_books"}}} Output JSON schema: {"type": "object", "required": ["books"], "properties": {"books": {"type": "array", "items": {"$ref": "#/book"}}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Email order**: Send the order confirmation and the invoice to the customer. If neither Resend nor SMTP is configured, the action still succeeds and the status is exactly "not_sent: email provider not configured". Pay already attempts this once. Input JSON schema: {"type": "object", "required": ["action", "order_id"], "properties": {"action": {"const": "notify_order"}, "order_id": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["notification"], "properties": {"notification": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Pay (sandbox)**: SANDBOX payment simulator, not a live charge. 4242424242424242 succeeds (Visa). 4000000000000002 declines. The full card number is never stored or returned; only brand and last4. Set STRIPE_SECRET_KEY to send the same call to Stripe test mode instead. A successful sandbox charge creates the invoice and attempts the confirmation email. Input JSON schema: {"type": "object", "required": ["action", "order_id", "card_number"], "properties": {"action": {"const": "pay"}, "order_id": {"type": "string"}, "card_number": {"type": "string"}, "exp_month": {"type": "integer"}, "exp_year": {"type": "integer"}, "cvc": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["order", "payment"], "properties": {"order": {"type": "object"}, "payment": {"type": "object"}, "invoice": {"type": "object"}, "notification": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Register customer**: Create a customer. Email must contain a name, a domain, and a dot. tax_id is optional and is not checked against a registry. Input JSON schema: {"type": "object", "required": ["action", "name", "email", "billing"], "properties": {"action": {"const": "register_customer"}, "name": {"type": "string"}, "email": {"type": "string"}, "tax_id": {"type": ["string", "null"]}, "billing": {"type": "object", "required": ["line1", "city", "postal_code", "country"], "properties": {"line1": {"type": "string"}, "line2": {"type": ["string", "null"]}, "city": {"type": "string"}, "region": {"type": ["string", "null"]}, "postal_code": {"type": "string"}, "country": {"type": "string"}}}}} Output JSON schema: {"type": "object", "required": ["customer"], "properties": {"customer": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **Search books**: Case-insensitive match on title or author. Input JSON schema: {"type": "object", "required": ["action", "query"], "properties": {"action": {"const": "search_books"}, "query": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["books"], "properties": {"books": {"type": "array"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.
- **View cart**: Return the open cart and line totals. Input JSON schema: {"type": "object", "required": ["action"], "properties": {"action": {"const": "view_cart"}, "cart_id": {"type": "string"}, "context_id": {"type": "string"}}} Output JSON schema: {"type": "object", "required": ["cart"], "properties": {"cart": {"type": "object"}}} Send this as an A2A DataPart for a deterministic result that does not call a model. A TextPart is handled by the language model, which calls the same tool. Every result also includes state and message.

## Call it over A2A

```bash
curl -X POST "https://agent.agiq.top" \
  -H "content-type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"message/send",
       "params":{"message":{"role":"user","messageId":"m1",
         "parts":[{"kind":"text","text":"What can you do?"}]}}}'
```