Deployment & infraDatabases & storageAI models

Cloudflare

Edge compute, DNS, storage, workers, AI, and official MCP servers.

Agents can deploy Workers, manage DNS, inspect analytics, provision storage, and connect through Cloudflare MCP servers.

Agent access evidence

CLI

native

Wrangler CLI manages Workers, Pages, D1, KV, R2, and local dev.

Evidence

API

native

Cloudflare API reference covers account, DNS, Workers, Pages, R2, KV, and security products.

Evidence

MCP

native

Official Cloudflare MCP server exposes the Cloudflare API through token-efficient Code Mode; Cloudflare also publishes a docs MCP server.

Evidence

Browser

strong

Dashboard is important for account, DNS, logs, and billing review.

Evidence

Account creation

strong

Self-serve signup.

Evidence

Pricing clarity

strong

Public pricing across plans and usage-based developer products.

Evidence

Docs quality

strong

Official docs are listed for API behavior, authentication, limits, examples, and setup review.

Evidence

Sandbox

strong

Local dev, preview deployments, and scoped API tokens support safer automation.

Evidence

Readiness score

Strong

Strong: 81/100. Strongest in machine operability; weakest in production reliability.

Machine operability24/25

Agent safety20.4/25

Agent readability17.4/20

Auth and setup13/15

Production reliability6.2/15

Best for

Edge deployment, DNS, Workers, storage, observability, and agent infrastructure.

Pricing

Public pricing across plans and usage-based developer products.

Auth

API tokens with scoped permissions, OAuth, account IDs, and Wrangler login.

Account creation

Self-serve signup.

Browser support

Dashboard is important for account, DNS, logs, and billing review.

CLI

wrangler

API

https://api.cloudflare.com/client/v4

MCP

https://mcp.cloudflare.com/mcp

Limitations

Key limitations: DNS, firewall, and production deployment changes need explicit human approval. Model behavior, costs, rate limits, data retention, and safety policies can change by model and account tier; use evals, spend caps, and data controls for sensitive workloads. MCP servers inherit the connected account's scopes and may send prompts or data to third-party services; configure read-only or narrow scopes when available. CLI access is powerful but can run destructive local or production commands; prefer dry runs, preview environments, and non-production credentials.