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.

DimensionAzure DevOps MCP ServerBring Azure DevOps projects, work items, repos, wikis, and pipelines to your AI agentMartinLoop MCP ServerAn independent control layer that gives AI coding agents budgets, brakes, and receipts
FMRS64 / 100 · C79 / 100 · B
Reliability7 / 2012 / 20
Security and permissions12 / 2018 / 20
Maintenance16 / 2016 / 20
Documentation15 / 2018 / 20
Setup experience14 / 2015 / 20
Best for
  • Teams already using Azure DevOps for work items, repos, wikis, or pipelines
  • Developers using VS Code with GitHub Copilot (or another MCP-capable client)
  • Scenarios wanting natural-language access to Azure DevOps without hand-writing REST API calls
  • 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
  • Teams not using Azure DevOps, or relying solely on GitHub or other project-management tools
  • Users wanting a minimal, zero-configuration server with no domain selection or account sign-in
  • Scenarios highly sensitive to breaking changes during preview — the README notes a recent full tool renaming/consolidation
  • 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 OAuth browser sign-in with a Microsoft account matching the target Azure DevOps organization (remote and local)
  • Local server also supports authentication via --authentication azcli
  • Requires an Azure DevOps organization identifier (e.g. 'contoso')
  • Local mode can set default project and team via environment variables to skip selection prompts
  • 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
  • Broad tool surface may allow modifying organizational data such as wiki pages and work items — verify the signed-in account's permission scope
  • The remote server is in preview, with functionality and issue tracking still evolving
  • A recent full tool consolidation renamed existing tools, which can break existing agents/skills unless pinned to an older version or updated per docs/TOOLSET.md
  • The local server requires Node.js 20+; an unmet version requirement will cause install or runtime failures
  • 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 clientsVS CodeClaude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue
Tools00