whatsapp-mcp

An MCP server that bridges an AI client (e.g., Claude Desktop/Cursor) to WhatsApp by letting the agent search contacts and chats, read message history from a local SQLite store, and send text and media via a locally running WhatsApp “bridge” (Go) and webhook forwarding for incoming messages.

Evaluated Apr 04, 2026 (40d ago)
Repo ↗ Communication mcp model-context-protocol whatsapp agent-tools go fastmcp sqlite webhooks media
⚙ Agent Friendliness
54
/ 100
Can an agent use this?
🔒 Security
42
/ 100
Is it safe for agents?
⚡ Reliability
31
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
70
Documentation
75
Error Messages
0
Auth Simplicity
55
Rate Limits
10

🔒 Security

TLS Enforcement
60
Auth Strength
35
Scope Granularity
10
Dep. Hygiene
45
Secret Handling
60

Security concerns: explicit prompt-injection/exfiltration risk is called out. The system stores WhatsApp messages locally in SQLite and can forward incoming messages to an external webhook, increasing data handling risk. App-level authorization/scoped tool permissions are not described; WhatsApp authentication is via QR-based session, but access control for who can call MCP tools appears to rely on local deployment/network controls. TLS and detailed error handling, auth headers, and rate limiting are not described in the README.

⚡ Reliability

Uptime/SLA
0
Version Stability
45
Breaking Changes
35
Error Recovery
45
AF Security Reliability

Best When

You run it locally/single-user, want agent-driven access to your own WhatsApp data, and can control network exposure to the bridge/webhook endpoints.

Avoid When

You cannot safely handle sensitive messaging data (PII/secrets), or you need strict safeguards against prompt injection/data exfiltration and unauthorized tool use.

Use Cases

  • Agent-assisted WhatsApp support: look up recent messages and draft replies
  • Automated internal workflows that read WhatsApp messages and forward them to an external system via webhook
  • Contact discovery: resolve names from phone numbers/JIDs
  • Lightweight personal assistant: retrieve message context and send text/media to known contacts/groups

Not For

  • Production-grade multi-tenant deployments without additional access controls
  • Use in environments where storing WhatsApp message content locally is disallowed
  • High-assurance compliance workflows without further security review and audit

Interface

REST API
Yes
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
Yes

Authentication

Methods: WhatsApp Web QR login via WhatsApp Web session
OAuth: No Scopes: No

No user-facing OAuth/scoped authorization is described for the MCP tools; access appears to be controlled by local process/network exposure and the WhatsApp session created via QR authentication.

Pricing

Free tier: No
Requires CC: No

Open-source project; cost is infrastructure/compute and any external webhook receiver costs.

Agent Metadata

Pagination
limit-based
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Potential prompt-injection risk when the agent is allowed to read/send sensitive message content (explicitly warned in README).
  • Media downloads/sends may require correct message_id/chat_jid context; the agent must supply both for download_media.
  • Local SQLite storage means agents should be constrained to only the data they are authorized to access (README mentions “only sent to Claude when you allow it”, but enforcement details are not specified).
  • Bridge is a local service; if exposed on a network, tool access may be reachable without strong app-level auth (not described).

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for whatsapp-mcp.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-04-04.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered