M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
COMPARE UP TO 4 SERVERS

Compare MCP servers

Compare scores, permissions, risks, and fit in one decision-focused table.

DimensionFlyto2 CoreOpen-source execution engine for AI agents with 468 modules, MCP-native, triggers, queue, versioning, and metering.MartinLoop MCP ServerAn independent control layer that gives AI coding agents budgets, brakes, and receipts
FMRS50 / 100 · D79 / 100 · B
Reliability8 / 2012 / 20
Security and permissions8 / 2018 / 20
Maintenance12 / 2016 / 20
Documentation12 / 2018 / 20
Setup experience10 / 2015 / 20
Best for
  • Developers needing reliable, replayable workflow automation
  • Teams wanting deterministic tools for AI agents
  • Users requiring unified execution engine for browser, API, and file operations
  • Engineering teams scaling AI coding agents while controlling spend
  • Platform teams that need auditable, rollback-aware agent execution records
  • Organizations using multiple agents (Claude, Codex, Gemini) under one governed flow
Not for
  • Users looking for a hosted cloud service rather than self-hosting
  • Users needing non-Python language support
  • Users needing simple scripts without advanced replay and evidence features
  • Developers who just want a simple code-completion helper without run governance
  • Teams unwilling to accept local JSONL run records with HMAC signing
  • Environments without a supported adapter (Claude/Codex/Gemini CLIs, direct provider, or verifier-only)
Required permissions
  • File system read/write (sandboxed path guard)
  • Network requests (guarded HTTP clients)
  • Environment variable reads (93 documented readers)
  • Browser automation via Playwright
  • Read/write access to the configured repo root (writable scope restrictable via --allow-path/--deny-path)
  • Ability to invoke the chosen agent CLI (Claude/Codex/Gemini, etc.) and its underlying model account
  • Local disk writes for run records, receipts, and share outputs (share/run-receipt., etc.)
  • Execution of user-configured verifier commands (e.g. npm test)
Risks and side effects
  • Sandbox path guard may block some file writes
  • Network access may be restricted, affecting external API calls
  • Requires Python environment for MCP server
  • Active development may change features
  • Verifier commands are executed; poorly chosen verifier commands carry security risk despite preflight policy checks
  • Cost and token figures may be estimates (labeled with provenance) and should not be treated as settled accounting
  • Receipt integrity must be verified before use as trustworthy evidence; missing rollback evidence is flagged as EVIDENCE_BOUNDARY
  • Policy defaults come from martin.config.yaml and can be overridden by CLI flags; misconfiguration can loosen safety boundaries
Supported clientsClaude DesktopClaude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue
Tools10