OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

Wappy

Wappy is an open-source TypeScript SDK and project generator for building AI agents that communicate through WhatsApp and run on a developer's own machine. It combines an agent harness with WhatsApp integration and supports developer-selected models and memory backends.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Wappy — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Wappy when you want to build and run a custom WhatsApp agent on your own machine, and can configure the WhatsApp Cloud API and the model or local runtime it depends on.

Who it is for

  • Developers building personal assistants or chatbots on WhatsApp.
  • Teams creating custom support, booking or internal operations agents and able to connect their own systems and implement the business logic.

Documented uses

  • Generate a TypeScript agent project for WhatsApp.
  • Choose an AI model provider or use local Ollama.
  • Store and retrieve ingested knowledge using local SQLite/LibSQL, or connect an optional Cognee backend.
  • Add read-only Gmail and Calendar search tools using the optional productivity and Google connector packages.
Before you choose

Requirements & limits

The project is open source under the MIT license. Its README says no Wappy account is required. A WhatsApp Cloud API app and credentials are required; model-provider API credentials are needed unless using local Ollama.

  • Requires Node.js 22 or newer and a configured WhatsApp Cloud API app.
  • Supports Anthropic, OpenAI and Gemini model providers, as well as local Ollama.
  • The default SQLite/LibSQL memory option is local; Cognee is an optional alternative, with Python required only for self-hosting it without Docker.
  • The Gmail and Calendar example uses read-only tools and requires the optional productivity and Google connector packages.
  • Wappy supplies the framework, not completed business integrations: developers must connect their own systems and implement custom workflows.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a Wappy agent project

Generate a project and configure its model and WhatsApp credentials.

Before you start

  • Node.js 22 or newer
  • A WhatsApp Cloud API app, phone number, access token and app secret
  • A model API key for Anthropic, OpenAI or Gemini, or local Ollama

Steps

  1. Run the command below and answer the setup questions, including model provider and knowledge-storage choice.

    Documented command
    npm create @wappy_ai/agent

  2. Copy .env.sample to .env and fill in the model key and WhatsApp credentials.

  3. Run the command below.

    Documented command
    npm install

  4. Run the command below; the command starts the server, opens a tunnel and prints a URL for Meta's webhook configuration.

    Documented command
    npm run dev

Expected outcome

A generated agent project with its development server started and a webhook URL available for Meta configuration.

Guide limitations

  • The guide documents project setup and server startup, not completion of Meta webhook configuration or a verified live WhatsApp conversation.
  • Business workflows and connections to a user's own data and tools must be implemented by the developer.
  • Cognee self-hosted without Docker requires Python 3.10–3.14; the default local memory backend does not.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
10 October 2026
Original website title
csr1010/wappy-kit · Official homepage title · retained in its source language
How we review sourcesSuggest a correction