Claude Skill

cloudflare

Run every Cloudflare job through the official `cf` CLI - domains, DNS, Workers deploys, secrets and custom domains, KV, D1, R2, WAF, SSL and cache. Use when the user mentions Cloudflare, cf, DNS, a domain, deploy, Workers, KV, D1, R2, CDN, SSL or cache.

LLM Mart · 0 points · 2 views 0 listing impressions 0 install-command copies
Download melvynx-ai-maxxing-skills-skills_cloudflare-4c7f30a.zip · 1 KB

Install

skills CLI npx skills add https://github.com/Melvynx/ai-maxxing-skills/tree/main/skills/cloudflare
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install melvynx-ai-maxxing-skills@llmmart
Git git clone https://github.com/Melvynx/ai-maxxing-skills.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole melvynx/ai-maxxing-skills collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

Use the official cf CLI (npm i -g cf) for every Cloudflare action. Never Wrangler, an S3 SDK script or raw calls to api.cloudflare.com; if cf cannot do something, say so and stop. The Cloudflare connector in Claude can help for reads, but changes go through cf.

cf --version                 # missing: run the setup-cloudflare skill
cf auth whoami               # must say "authenticated": true
cf auth login                # OAuth in the browser, the user grants every scope, sensitive ones included

When cf stops with "More than one account available", run cf accounts list and export CLOUDFLARE_ACCOUNT_ID=<id> for the session. Zone commands take -z <domain or zone id>. Never print tokens or read the cf config file.

Find any command instead of guessing it, with generic queries (action + resource type, no names or ids):

cf cli search "<what you want to do>"     # JSON matches
cf <discovered command> --help              # flags of that command
cf schema <command without the leading cf>  # exact API request body

The commands used most:

cf zones list
cf dns records list -z <domain>
cf dns records create -z <domain> --body '{"type":"CNAME","name":"www","content":"target","proxied":true}'
cf init <dir> && cf dev        # then cf build, cf deploy
cf r2 buckets list
cf r2 objects put <key> --bucket-name <bucket> --file <path> --content-type <mime>

Everything else (DNS update or delete, Worker custom domain, secrets, routes, crons, KV, D1, cache purge, SSL, WAF, Access, email routing, DNSSEC, analytics) goes through cf cli search "<job>". Add --local to simulate KV, D1 and R2 objects. Subcommand names follow the CLI release: if one errors, search it.

Every mutating command accepts --dry-run. Confirm with the user before anything destructive (DNS deletes, zone changes, purges), and pass --force only after their yes.

When another service (Lumail, Stripe, Google, a mail provider) shows DNS records to add, create each one exactly as given (type, name, content, priority, TTL), keep mail and verification records unproxied ("proxied": false), then list the zone to confirm they exist.

Files (ai-maxxing-skills)
  • SKILL.md 2.4 KB
    ---
    name: cloudflare
    description: Run every Cloudflare job through the official `cf` CLI - domains, DNS, Workers deploys, secrets and custom domains, KV, D1, R2, WAF, SSL and cache. Use when the user mentions Cloudflare, cf, DNS, a domain, deploy, Workers, KV, D1, R2, CDN, SSL or cache.
    ---
    
    Use the official `cf` CLI (`npm i -g cf`) for every Cloudflare action. Never Wrangler, an S3 SDK script or raw calls to api.cloudflare.com; if `cf` cannot do something, say so and stop. The Cloudflare connector in Claude can help for reads, but changes go through `cf`.
    
    ```bash
    cf --version                 # missing: run the setup-cloudflare skill
    cf auth whoami               # must say "authenticated": true
    cf auth login                # OAuth in the browser, the user grants every scope, sensitive ones included
    ```
    
    When `cf` stops with "More than one account available", run `cf accounts list` and `export CLOUDFLARE_ACCOUNT_ID=<id>` for the session. Zone commands take `-z <domain or zone id>`. Never print tokens or read the `cf` config file.
    
    Find any command instead of guessing it, with generic queries (action + resource type, no names or ids):
    
    ```bash
    cf cli search "<what you want to do>"     # JSON matches
    cf <discovered command> --help              # flags of that command
    cf schema <command without the leading cf>  # exact API request body
    ```
    
    The commands used most:
    
    ```bash
    cf zones list
    cf dns records list -z <domain>
    cf dns records create -z <domain> --body '{"type":"CNAME","name":"www","content":"target","proxied":true}'
    cf init <dir> && cf dev        # then cf build, cf deploy
    cf r2 buckets list
    cf r2 objects put <key> --bucket-name <bucket> --file <path> --content-type <mime>
    ```
    
    Everything else (DNS update or delete, Worker custom domain, secrets, routes, crons, KV, D1, cache purge, SSL, WAF, Access, email routing, DNSSEC, analytics) goes through `cf cli search "<job>"`. Add `--local` to simulate KV, D1 and R2 objects. Subcommand names follow the CLI release: if one errors, search it.
    
    Every mutating command accepts `--dry-run`. Confirm with the user before anything destructive (DNS deletes, zone changes, purges), and pass `--force` only after their yes.
    
    When another service (Lumail, Stripe, Google, a mail provider) shows DNS records to add, create each one exactly as given (type, name, content, priority, TTL), keep mail and verification records unproxied (`"proxied": false`), then list the zone to confirm they exist.
    

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.