blog
How Claude Code actually works under the hood — transcripts, skills, customization layers — and what to do with that knowledge.
Attribute AI coding spend from Claude Code, Cursor, Codex, and Gemini CLI to each developer, repository, pull request, and client. The options per tool, what each one misses, and a setup that covers all of them.
Track what your team spends on Cursor per developer, repository, pull request, and client. What Cursor's team dashboard covers, what it leaves out, and how to fill the gap with budgets and alerts.
Use Claude Code's built-in OpenTelemetry export to track cost per developer and repository. The settings, the api_request event fields, the gotchas (no default protocol, repo attributes off by default), and where to send it.
See which repositories and models drive your Cursor bill: Cursor's usage dashboard and APIs compared with local, per-request tracking tagged by git repo.
How to see and check Claude Code token usage, plan limits, and session cost with /usage (/cost, /stats), a status line, /context, ccusage, the Console, and PromptConduit.
Step-by-step guide to writing your first Claude Code skill: the SKILL.md format, frontmatter fields, when to use project vs user scope, and a complete /review skill you can copy.
A complete guide to Claude Code customization: CLAUDE.md, Skills, slash commands, hooks, and memory. Where each one lives, when to use it, and how they overlap.
Claude Code saves each session as JSONL in ~/.claude/projects (%USERPROFILE%\.claude\projects on Windows) and deletes them after 30 days by default.