The MCP Server Directory
Every Model Context Protocol server, scored on reliability, documentation, maintenance, and adoption. Independent, opinionated, refreshed daily from the official registry.
What this is
The MCP Server Directory is a live, independently maintained index of Model Context Protocol servers published to the official registry. Each server is scored 0–10 on four dimensions: reliability, documentation quality, maintenance velocity, and adoption. The directory is free, updated daily, and built by MegaOne AI.
Blackpoint MCP
4.3/10MCP server for Blackpoint Cyber MDR — alerts, threats, and operations.
Datto Rmm MCP
4.3/10MCP server for Datto RMM — devices, alerts, sites, jobs, and audit data.
Kaseya Bms MCP
4.3/10MCP server for Kaseya BMS PSA — tickets, accounts, time entries, contracts.
Kaseya Vsa MCP
4.3/10MCP server for Kaseya VSA RMM — agents, patches, procedures, alarms, tickets.
Ninjaone MCP
4.3/10MCP server for NinjaOne RMM — devices, organizations, alerts, and tickets.
Salesbuildr MCP
4.3/10MCP server for SalesBuildr — quoting and sales workflows for MSPs.
Sentinelone MCP
4.3/10Multitenant Streamable HTTP wrapper for the SentinelOne purple-mcp server.
Superops MCP
4.3/10MCP server for SuperOps unified PSA+RMM — clients, tickets, assets, and more.
Syncro MCP
4.3/10MCP server for Syncro MSP — customers, tickets, assets, invoices, and more.
Cap Shield
4.3/10Context selection and compression for AI agents - recall measured, not claimed.
Octowatch MCP
4.3/10Read-only OctoWatch DLP Cloud MCP: risks, idle time, productivity, monitoring.
Agent Residue Evidence
4.3/10Report local task-scoped test and build residue without cleaning resources.
Agent Runtime Proof
4.3/10Verify whether a local Agent or MCP runtime matches an approved artifact.
MCP Supertoinette
4.3/10Read Supertoinette recipes and rescale their quantities honestly. No API key.
Sippion
4.3/10Local read-only repository context retrieval for AI coding agents.
Kmp
4.3/10Local-first agent memory that preserves what happened, when and why
MCP Server
4.3/10Stdio bridge to Sourcey startup offers and Agent Readiness report cards.
Phi Guard MCP
4.3/10Local-first PHI detection for source code, LLM prompts, logs, and analytics
Agent Spreadsheet
4.3/10Agent-safe Excel workbook analysis, editing, recalc, and verification tools.
Deepnatal
4.3/10Natal charts cross-verified three ways. Refuses to answer rather than guess.
Citable MCP
4.3/10SEO and AI-visibility checks agents buy per call — USDC on Solana, no API key.
Cloudarch Codeflow
4.3/10Turn a TypeScript codebase into a live animated call-flow diagram on CloudArch.
Platform MCP
4.3/10Read-only GCP tools for debugging incidents and finding cost savings
Agent Recall Trail
4.3/10Private Agent memory and governed shared knowledge.
Arc Control MCP
4.3/10Arc browser MCP for macOS: agent window, snapshots, DevTools input, guardrails
Sinter
4.3/10Persistent, evidence-backed code graph for coding agents.
Ictcrm MCP
4.3/10Read CRM contact groups and manage contacts on ICTCRM (ICTContact backend).
MCP Apple Safari
4.3/10Read and act on Safari — history, tabs and reading list, writes off by default
Assistantmail MCP
4.3/10Agent email with allowlist and consent via AssistantMail.
Memdsl
4.3/10Review-gated declarative long-term memory with bounded read-only MCP access.
Fidelis Memory
4.3/10Recall locally stored agent memories through an MCP-compatible host.
Rpfd
4.3/10Read and write RAGE Package Format archives
Obelisk
4.3/10Comment on a passage of an Obsidian note, and propose suggested edits.
Cyberwarex Chain Data
4.3/10EVM onchain data: wallet, token, price, gas, transactions, ENS across 5 chains.
Cyberwarex Voice Stt
4.3/10Speech-to-text transcription for AI agents.
Cyberwarex Web Access
4.3/10Web access for agents: fetch, extract, screenshot, or PDF any public URL.
Github Projectpulse MCP
4.3/10Monitor GitHub repo health, DORA metrics and CI signals from your AI assistant.
Promptready MCP
4.3/10Convert PDF/CSV to Markdown with AI-powered OCR via the PromptReady cloud API.
Agent Browser
4.3/10Self-hosted Chrome for AI agents, with a live noVNC view a human can take over.
Memoryguard
4.3/10Local-first governed MCP memory backend for coding agents.
Stripe MCP
4.3/10Read-only MCP server for querying Stripe customers, payments, and invoices.
Neither MCP
4.3/10Project context for Cursor/Claude via stdio MCP. Key: www.neither.online/start
Ryrank
4.3/10Cohere-compatible pay-per-call reranker, no signup, x402/USDC
Snap Router
4.3/10Task-to-tool discovery over the merged x402 Bazaar + MCP Registry catalog
Deposit
4.3/10Take a client deposit, hold it, set it against an invoice or give it back.
Quartermaster
4.3/10Local MCP asset intelligence across Unity, Fab, Megascans, Gumroad & Cosmos.
Abcop
4.3/10ABC complexity gate for AI-written code with an MCP inspection tool.
Free · No spam
New servers, score changes, breakages
Every Monday: what shipped, what broke, what's worth trying. Five minutes. No filler.
Four dimensions, weighted
- Reliability (30%) — repository present, permissive license, not archived, official status.
- Documentation (20%) — description length, website presence, README quality.
- Maintenance (25%) — how recently the code was pushed; versioning discipline.
- Adoption (25%) — GitHub stars (log-scaled), remote availability, official endorsement.
What is MCP?
MCP stands for Model Context Protocol. It is an open standard introduced by Anthropic in late 2024 that lets AI assistants like Claude connect to external tools, data sources, and services through a common interface. An MCP server is a small program that exposes one capability (file access, a database, an API) to any MCP-compatible AI client.
Where does the data come from?
Every server in this directory is pulled from the official Model Context Protocol Registry at registry.modelcontextprotocol.io. We also enrich each entry with GitHub metadata (stars, last commit, license, primary language) where a repository is linked. The full re-scan runs daily.
How are servers scored?
Each server gets a 0–10 overall score, weighted from four sub-scores: reliability (30%), documentation (20%), maintenance (25%), and adoption (25%). The full methodology is visible on this page under "How scoring works".
Is this affiliated with Anthropic?
No. MegaOne AI is independent. Anthropic created the MCP standard and maintains the reference implementations, but this directory is built separately and rates all servers including Anthropic's own.
How often does the directory update?
The pipeline re-fetches the full registry and re-scores every active server once per day. Individual entries are refreshed whenever a new version is published to the official registry.
Is it free?
Yes. Browsing, filtering, and searching the directory is free. The weekly email briefing is free. There is no account required.