{"id":"delltrak-wamcp","name":"wamcp","af_score":71.0,"security_score":54.8,"reliability_score":43.8,"what_it_does":"WA MCP is a Model Context Protocol (MCP) server that exposes WhatsApp functionality to MCP-compatible AI agents. It provides 63 tools (prefixed with wa_), 10 read-only resources (whatsapp:// URIs), and 12 real-time events (SSE) for both WhatsApp Web via Baileys and WhatsApp Cloud API via a dual-channel backend. It can be deployed with Docker using Redis and supports multi-instance WhatsApp connections with queueing and persistent storage.","best_when":"You want an MCP-native bridge so an agent can discover and call WhatsApp tools automatically, and you can run/secure a long-lived MCP server with Redis and WhatsApp credentials.","avoid_when":"You cannot securely manage WhatsApp session/credential material, or you lack operational support for reconnect/retry behavior and rate limiting.","last_evaluated":"2026-03-30T15:40:16.296639+00:00","has_mcp":true,"has_api":false,"auth_methods":["Baileys session/pairing via QR code or pairing code","Meta Cloud API token via wa_set_cloud_credentials"],"has_free_tier":false,"known_gotchas":["WhatsApp connectivity may require QR/pairing flows and may fail until instance is connected.","Rate limiting is queue-based; agents that fire many sends may still hit backlogs or receive errors during throttling.","Real-time events are delivered via SSE; ensure the agent runtime maintains the SSE connection.","Multi-instance setups require correct instance IDs and consistent routing of tool calls to the intended instance."],"error_quality":0.0}