Tools

Tools turn the assistant into something that does things: each is a function the AI can call mid-conversation — your webhook, or one Happ ships with.

Add a tool

1. Open the Tools tab inside an assistant.
2. Webhook — HTTP request to your API. Define name, description, params schema, URL, method, and a shared secret. Examples: create_reservation (POST), lookup_customer (GET).
3. Built-in — no setup, every assistant gets them: searchKnowledge (vector search over the attached knowledge base), turnOffAiMode (hands the conversation to a human operator with full context), searchPreviousConversations (looks up what this customer discussed earlier) and sendImage (sends a picture from the knowledge base into the chat).
4. Write descriptions that explain when to call, not just what the tool does — the AI sees name, description, and param schema in its prompt.
5. Hit Check on any tool to fire a test call and inspect the response.

The AI fills required params from context (asking the customer first if info is missing) and tool calls are recursive — results feed the next turn, so it can chain multiple tools in one reply.

Example webhook call

POST to your endpoint with collected params plus chatId:

{ "table_id": "T-12", "time": "2026-05-18T19:30:00Z", "guests": 4, "phone": "+380501234567", "chatId": "chat-uuid" }

Return short JSON the AI can paraphrase:

{ "result": "Booking confirmed. Confirmation number #12345." }

Give the tool a secret and every call carries Authorization: Basic <secret>. Calls with a body (POST, PUT, PATCH) also carry X-Signature: sha256=<hex> — HMAC-SHA256 of the raw body; recompute it before processing. A GET tool sends its params in the query string and carries no signature.

API: api.happ.tools/reference

Next

• Knowledge base — let it look up information before acting
• AI control — how turnOffAiMode hands off, and when AI re-enables
• Channels — where tool calls happen: Telegram, Instagram, WhatsApp, voice