Claude Code
Headless Claude Code: auth, limits, logging, routing.
Claude Code slash commands, custom commands, and how skills fit in
Which Claude Code slash commands are worth learning, how custom commands relate to skills, and why a command can fail to show up or run.
Claude Code Hooks: Events, Input JSON, Exit Codes, Examples
How Claude Code hooks work: where to configure them, the JSON they receive, exit code 2 vs 0, matchers, working examples, and why a hook may not be running.
Where the CLAUDE.md file lives and what to put in it
Where Claude Code looks for a CLAUDE.md file, what belongs in it, how imports and scopes work, and copy-paste examples for real projects.
Using claude setup-token on a server without leaking it
How to run claude setup-token, store the long-lived token in a systemd EnvironmentFile, keep it out of argv and logs, and stop API key billing.
Claude Code headless: running claude -p in a production worker
How to run Claude Code headless with claude -p on a small VPS: OAuth token auth, systemd env, tool permissions, review runs, and real wall-clock times.
Claude Code usage limit reset: parking unattended jobs
What a Claude Max usage limit does to an unattended Claude Code worker, how the reset works, and how to park jobs as QUOTA_WAIT and resume them.
Claude Code output-format json: log tokens, cost and turns
How to use claude code --output-format json to capture tokens, cache reads, cost and turns per call, store them in SQLite, and handle usage limits.