Discover and explore top open-source AI tools and projects—updated daily.
matiasbattocchiaWhatsApp Business API platform for AI agents and multimodal data extraction
Top 85.8% on SourcePulse
OpenBSP API provides a self-hostable, multi-tenant platform for managing WhatsApp Business communications. Built on Deno and Supabase, it targets individual businesses and service providers aiming to offer WhatsApp messaging services, acting as a backend for AI agents and facilitating Meta Business Partner transitions. Its core benefit lies in offering a flexible, scalable, and integrated solution for WhatsApp API interactions and AI agent orchestration.
How It Works
The platform employs a reactive, function-based architecture centered around Supabase. Incoming WhatsApp messages are handled by the whatsapp-webhook Edge Function, which persists them in the Supabase database. Database triggers then route messages to the agent-client function, responsible for building conversation context and interacting with external AI agents via protocols like Chat Completions or A2A. Responses are saved back to the database, and a whatsapp-dispatcher Edge Function handles outbound message delivery. This design decouples communication logic from agent processing, promoting scalability and modularity.
Quick Start & Requirements
Self-hosting requires Deno and Docker. Deployment is streamlined via Supabase's GitHub Integration, allowing setup in under 15 minutes by forking the repository and connecting a Supabase project. A managed demo instance is available at web.openbsp.dev. Key setup involves configuring Meta App credentials and WhatsApp webhooks within Supabase Edge Functions.
Highlighted Details
Maintenance & Community
Community engagement is encouraged through a WhatsApp Community or by opening issues on GitHub. Specific details regarding core maintainers, sponsorships, or a public roadmap are not detailed in the provided README.
Licensing & Compatibility
The repository is described as open-source, but a specific license type (e.g., MIT, Apache 2.0) is not explicitly stated in the provided text. This lack of explicit licensing information may pose a compatibility concern for commercial use or integration into closed-source projects without further clarification.
Limitations & Caveats
Advanced AI agent logic is intended to be deployed as external services, requiring separate management. The setup for Meta App and WhatsApp API credentials involves several detailed steps across different Meta platforms. The README does not detail specific performance benchmarks or known bugs.
1 week ago
Inactive
danielgross
openclaw