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.

DimensionClaw OrchestratorUnified runtime to orchestrate Claude Code, Codex, Gemini, and other coding CLIs with multi-agent collaboration.MartinLoop MCP ServerAn independent control layer that gives AI coding agents budgets, brakes, and receipts
FMRS49 / 100 · D79 / 100 · B
Reliability6 / 2012 / 20
Security and permissions8 / 2018 / 20
Maintenance10 / 2016 / 20
Documentation13 / 2018 / 20
Setup experience12 / 2015 / 20
Best for
  • Developers and teams needing a unified interface to manage multiple coding agents.
  • Users who want to automate multi-agent workflows such as plan-code-review loops.
  • Individuals who wish to integrate MCP into their existing agent workflows.
  • Those interested in exploring agent orchestration or persistent session management.
  • 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
  • Simple use cases that do not require multi-agent orchestration or only use a single coding CLI.
  • Beginners preferring a full GUI without relying on external CLIs.
  • Environments that are conservative about non-official third-party tools.
  • 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
  • This server executes locally installed coding CLIs (e.g., `claude`, `codex`, `agy`, `agent`, `opencode`) as subprocesses.
  • It requires read/write access to working directories and git worktrees.
  • It may open network ports (default 18796) to serve a dashboard and HTTP API.
  • It may use the file system to store session state and configuration.
  • 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
  • As a third-party package, it is not officially certified by the upstream products, posing potential security risks.
  • Executing user-supplied agent commands can lead to arbitrary code execution; use in a controlled environment.
  • Dependence on external CLI versions may cause functionality issues due to version incompatibilities.
  • Network exposure (e.g., port 18796) could lead to unauthorized access; configure authentication if exposed.
  • 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 Desktop, Cursor, Zed, Windsurf, Continue, ClineClaude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue
Tools10