How to use Claude Code

Anthropic's agentic coding tool for the terminal. Below: 3 workflows, each with the steps and prompt.

Claude Code skills: build a /smoke-test skill that clicks through your web app in a real browser

A Claude Code skill is a folder with a SKILL.md file that Claude loads when your request matches its description, or that you run as /skill-name. To build a browser smoke test, install Playwright CLI and its agent skill, then add a project skill at .claude/skills/smoke-test/SKILL.md that lists your critical user flows and tells Claude to run them in a headless browser and report failures with screenshots and console errors. Type /smoke-test after a UI change, or name the skill verify (or simplify) so Claude Code runs it before each commit. In our 2026-10-15 test (Claude Code v2.1.300, Sonnet), the first run took about 30 s and $0.05 and found the planted bug without editing any code.

  1. Install Playwright CLI
  2. Install the Playwright agent skill into the repo
  3. Write down your critical flows
  4. Create the smoke-test skill
  5. Run it
  6. Triage and fix
  7. Optional: run it before every commit

n8n MCP server: build, test and publish n8n workflows from Claude Code

Enable Settings > Instance-level MCP in n8n, copy an MCP token from Connect > API key, and add the server to Claude Code with claude mcp add --transport http and an Authorization: Bearer header (the route we tested). Install the official n8n-skills plugin for n8n's conventions; its own OAuth connection is an untested alternative. Then ask Claude to plan, build, validate and test a workflow on pinned data. Publish it yourself after review, and block publish/archive tools with Claude Code deny rules. In our 2026-10-15 test (n8n 2.43.6, Claude Code v2.1.300) Claude built a 15-node unpublished workflow and passed a pinned-data test run in 166 s for $0.71, and the deny rules kept it from publishing or archiving anything.

  1. Enable MCP on the instance
  2. Connect Claude Code with an MCP token
  3. Add the n8n Skills plugin
  4. Add guardrails before the first prompt
  5. Ask for a plan first
  6. Let Claude build and validate
  7. Test on pinned data
  8. Review and publish yourself

Playwright MCP: set it up and have Claude Code test your web app

Add Microsoft's Playwright MCP server to your agent (for Claude Code: claude mcp add playwright npx @playwright/mcp@latest), point it at a local or staging URL, and ask it to walk a user flow step by step. It drives a real browser through accessibility snapshots, reports what broke, and can then write a regular Playwright test so the check runs in CI without the agent.

  1. Add the server to your agent
  2. Choose headed or headless
  3. Start your app and write the flow down
  4. Ask for a walkthrough, not a fix
  5. Check the evidence
  6. Fix, then re-run the same flow
  7. Turn it into a real test

Claude Code: pricing, features and all use cases