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.

DimensionMCP AppiumAppium MCP server for iOS and Android mobile automationMartinLoop MCP ServerAn independent control layer that gives AI coding agents budgets, brakes, and receipts
FMRS57 / 100 · C79 / 100 · B
Reliability9 / 2012 / 20
Security and permissions12 / 2018 / 20
Maintenance11 / 2016 / 20
Documentation14 / 2018 / 20
Setup experience11 / 2015 / 20
Best for
  • Teams doing Android/iOS automation testing with Appium
  • Developers who want to create and debug mobile tests in natural language
  • Engineering teams that need generated test code and Page Object templates
  • Teams running mobile automation in trusted CI environments
  • 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
  • Shared or multi-tenant services exposed to untrusted users
  • Environments without Android SDK, Xcode, or a Java toolchain
  • Pure browser or desktop web automation instead of mobile
  • 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
  • Read access to the local Android SDK and Xcode/simulator setup
  • Access to local mobile devices or emulators through adb or simulator tooling
  • Read/write access to the screenshots and recordings directory (SCREENSHOTS_DIR or the system temp directory)
  • Optional: network access to a remote Appium/WebDriver server (remoteServerUrl)
  • Optional: outbound calls to an external vision model API (AI_VISION_API_BASE_URL and AI_VISION_API_KEY)
  • 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
  • remoteServerUrl is caller-controlled; if it is derived from untrusted input it can reach unintended servers, so restrict destinations with REMOTE_SERVER_URL_ALLOW_REGEX
  • By default the server deletes all MCP-owned Appium sessions when the MCP client disconnects, and transports that drop easily such as httpStream can tear down automation in one go; use APPIUM_MCP_ON_CLIENT_DISCONNECT=skip to keep them
  • Enabling vision element finding sends screenshots to a third-party vision API, which can incur cost and data exposure
  • Persisted session files may contain credentials and sensitive capabilities (created with 0600 permissions)
  • Plugins execute inside the server process, so only trusted files and packages should be loaded
  • 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 clientsCursor, Claude Code, Gemini CLIClaude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue
Tools200