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.

DimensionAsk Grok MCP ServerConsult Grok through xAI's metered API with exact model selection and no fallbackMartinLoop MCP ServerAn independent control layer that gives AI coding agents budgets, brakes, and receipts
FMRS55 / 100 · C79 / 100 · B
Reliability9 / 2012 / 20
Security and permissions11 / 2018 / 20
Maintenance10 / 2016 / 20
Documentation12 / 2018 / 20
Setup experience13 / 2015 / 20
Best for
  • MCP client users who want an independent Grok 4.7 opinion
  • Users who need exact control over model ID, reasoning effort, and output-token ceiling and accept metered billing
  • Developers who already hold an xAI API key or have the official Grok CLI installed
  • 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 unwilling to pay metered API costs and looking only for free usage
  • Users needing automatic failover or multi-provider fallback — this server offers no harness or model fallback
  • Users who can neither provide XAI_API_KEY nor install grok-cli
  • Users who need fully offline or private local inference (a local provider would be a better fit)
  • 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
  • Requires the XAI_API_KEY secret environment variable, used only for requests to https://api.x.ai
  • If the grok-cli harness is used, requires an installed and authenticated official Grok CLI
  • Launched as a stdio process by the local MCP client from the npm package @ask-llm/grok-mcp
  • 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
  • The xAI API is metered and paid; prompt and response length directly affect cost, and ASK_GROK_MAX_OUTPUT_TOKENS only bounds accidental spend
  • XAI_API_KEY is a secret; if leaked, others could consume your quota
  • The server never falls back to another harness or model, so a misconfiguration causes failure rather than degradation
  • Grok's answers come from a third-party tool that is not an official xAI product; verify output accuracy yourself
  • Prompt content is sent to xAI's service, so treat sensitive code or data with care
  • 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 Code, Claude Desktop, Cursor, Codex, PiClaude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue
Tools10