| FMRS | 59 / 100 · C | 79 / 100 · B |
|---|
| Reliability | 8 / 20 | 12 / 20 |
|---|
| Security and permissions | 11 / 20 | 18 / 20 |
|---|
| Maintenance | 14 / 20 | 16 / 20 |
|---|
| Documentation | 13 / 20 | 18 / 20 |
|---|
| Setup experience | 13 / 20 | 15 / 20 |
|---|
| Best for | - Teams already using Cloudflare products who want to manage their account via natural language
- Workflows that benefit from curated, typed tools scoped to a single Cloudflare product area (observability, DNS, bindings, etc.)
- Users who want a remotely hosted MCP integration rather than a self-hosted/local server
| - 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 | - Workflows needing broad API coverage across many Cloudflare products via code execution (use Cloudflare's Code Mode server, cloudflare/mcp, instead)
- Users without a Cloudflare account, or who don't want to grant a third-party MCP client access to their account
- Scenarios requiring a fully offline or locally-run MCP server, since all listed servers are remotely hosted
| - 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 | - Optional Cloudflare API key passed as a Bearer token in the Authorization header, used when not authenticating via OAuth; supports user- or account-owned tokens
- OAuth authorization flow supported for connecting an account
- The Documentation server (docs.mcp.cloudflare.com) requires no auth header, serving public documentation lookups
| - 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 | - Each server maps to a different Cloudflare product domain; an overly broad API token scope could grant an MCP client more account access than a given task actually needs
- Some tools can make real changes to account configuration on the user's behalf, so misinterpreted natural-language instructions could cause unintended production changes
- Account credentials (API tokens) are transmitted as Bearer headers between the client and the remote service, so the client environment must be kept secure
- Some features may require a paid Cloudflare Workers plan
| - 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 clients | Claude, Cursor, OpenAI Responses API | Claude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue |
|---|
| Tools | 0 | 0 |
|---|