mcpbeat

mcpbeat

Uptime and reliability monitoring for MCP (Model Context Protocol) servers, tracking 26,000+ servers with live checks every 15 minutes instead of relying on GitHub star counts.

🔗 Visit mcpbeat
📁 Monitoring & Observability🗣️ English📅 September 1, 2026

Description

As more AI tools plug into external services through MCP (the protocol that lets AI agents call outside tools), a new problem showed up: how do you know if the MCP server you're about to integrate actually works, or if it's been abandoned and quietly stopped responding? mcpbeat answers that by actually pinging every MCP server it knows about, every 15 minutes, and publishing the results — real uptime, not just how many GitHub stars a project has.

mcpbeat maintains a monitoring database of 26,000+ MCP servers across 21 categories, performing live protocol handshakes every 15 minutes to verify responsiveness and detect when previously-exposed tools disappear. Its ranking algorithm weights responsiveness (40 points), weekly install activity (40 points), and exposed tool count (20 points) to produce a reliability score independent of popularity metrics like stars. The free tier lets anyone watch 3 servers with weekly email summaries; paid tiers ($20–$500/month) add unlimited monitoring, instant alerts via Slack/Discord/Teams, historical trend data, and auditor-ready reports, with enterprise API access starting at $500/month.

💬 Our review

The short version: mcpbeat fills a genuinely new gap — nobody was independently tracking whether MCP servers actually stay up — and does it with real, frequent live checks rather than a popularity proxy.

General uptime monitoring (UptimeRobot, Pingdom, Better Uptime) has existed for years but isn't purpose-built for the MCP handshake protocol or for surfacing tool-level changes (a server staying "up" while quietly removing the tool you depend on). mcpbeat's differentiation is protocol-specific: it verifies the actual MCP handshake, not just an HTTP 200, and tracks exposed-tool changes over time, which a generic uptime checker won't catch. The tradeoff is that it's a young, single-purpose tool in a fast-moving ecosystem — MCP itself is young enough that server churn is high, and mcpbeat's own longevity and data history are correspondingly short. Worth using free (3 servers) to sanity-check any MCP server before you build a dependency on it; upgrade to a paid tier only once you're managing enough MCP integrations that manual spot-checking becomes real operational risk.

💰 Pricing

FreemiumGratuit 3 serveurs, Pro 20$/mois
Free $0Pro $20/monthTeam $200/monthBusiness $500/month

📊 Global score

58Average
🌐Availability15/100Faible

1 language · 0 platform

📄Profile100/100Excellent

Profile completeness

🤖 AI-enriched data

💰 Pricing model
🆓 Freemium

Gratuit pour 3 serveurs surveillés ; Pro à 20 $/mois (illimité), Team 200 $/mois, Business 500 $/mois, accès API entreprise à partir de 500 $/mois

👥 Target audienceDéveloppeurs et équipes IA intégrant des serveurs MCP, mainteneurs de serveurs MCP cherchant des métriques d'adoption
🗣️ Languagesen
🌍 Target countriesInternational
👍

Pros

Vérification réelle du handshake MCP, pas juste un ping HTTP

Base de 26 000+ serveurs surveillés toutes les 15 minutes

Détection de disparition d'outils exposés dans le temps

Palier gratuit utilisable (3 serveurs)

👎

Cons

Écosystème MCP jeune et changeant, historique de données limité

Outil mono-fonction, pas un remplaçant d'un monitoring généraliste

Paliers payants élevés pour un usage équipe (200-500 $/mois)

❓ Frequently asked questions

What is mcpbeat in one sentence?
Is it free?
What exactly does it check?
How is the reliability score calculated?
Can I get alerts in Slack or Discord?
Is it worth the money compared to alternatives?
Which tool should you pick for your case?