# Claude Code Camp

> Weekly lessons on shipping faster with AI

This file provides information about Claude Code Camp to help large language models understand and reference this publication's content.

## Posts
- [I tried to build my own Ultrafast model](https://www.claudecodecamp.com/p/i-tried-to-build-my-own-ultrafast-model): Trying to replicate OpenAI's ultrafast using Qwen3-235B
- [Catch-Up Thursday](https://www.claudecodecamp.com/p/catch-up-thursday-oct-1): Everything that happened in AI coding this week
- [How models are served and how that impacts your bill](https://www.claudecodecamp.com/p/how-models-are-served-and-how-that-impacts-your-bill): I rented a GPU and served an open model to understand how prices work
- [Why new models get cheaper](https://www.claudecodecamp.com/p/why-new-models-get-cheaper): Opus 5.5 and GPT-6 Sol cost about half as much per task.
- [I moved my coding agents to a remote server](https://www.claudecodecamp.com/p/i-moved-my-coding-agents-to-a-remote-server): My laptop is no longer the bottleneck
- [Which model to use?](https://www.claudecodecamp.com/p/which-model-to-use): Fable, Sonnet, Opus and now GPT 5.6. Which one to use?
- [Someone put ads in the Claude Code spinner](https://www.claudecodecamp.com/p/someone-put-ads-in-the-claude-code-spinner): Plus: understanding AI code, and how /fast works
- [How to Use Claude Code and Still Understand Your Own Code](https://www.claudecodecamp.com/p/how-to-use-claude-code-and-still-understand-your-own-code)
- [How Claude Code Fast Mode Works](https://www.claudecodecamp.com/p/how-claude-code-fast-mode-works): Why it's 2.5× faster, why it costs 2–6× more, and when to turn it on.
- [Fable 5 is offline. The government pulled it.](https://www.claudecodecamp.com/p/fable-5-is-offline-the-government-pulled-it): Here's what we know.
- [Two weeks with Claude Fable 5](https://www.claudecodecamp.com/p/two-weeks-with-claude-fable-5): and what is Loop Engineering?
- [Images cost 3x more tokens in Claude Opus 4.7](https://www.claudecodecamp.com/p/images-cost-3x-more-tokens-in-claude-opus-4-7): Here's what image tokenization is, what changed in 4.7, and when it's actually worth paying for
- [I Measured Claude 4.7's New Tokenizer. Here's What It Costs You.](https://www.claudecodecamp.com/p/i-measured-claude-4-7-s-new-tokenizer-here-s-what-it-costs-you): The docs said 1.0–1.35x more tokens. On real content, I measured 1.47x.
- [Claude Code Sandbox: How /sandbox Works (and What It Doesn't Protect)](https://www.claudecodecamp.com/p/claude-code-sandboxing-how-sandbox-works-and-what-it-doesn-t-protect): What Claude Code's /sandbox blocks at the OS, what still slips through, and a settings.json that matches the experiments
- [Claude Code Pricing (2026): Plans, Max vs API, and How Much It Costs](https://www.claudecodecamp.com/p/claude-code-pricing): Pro, Max, Team, and API rates, plus real session math with prompt caching
- [I Tried to Reverse Engineer Claude Code's Usage Limits](https://www.claudecodecamp.com/p/i-tried-to-reverse-engineer-claude-code-s-usage-limits): Anthropic sends your utilization data on every API response. Claude Code doesn't show it to you. So I built a proxy to catch it.
- [Claude Code Extended Thinking](https://www.claudecodecamp.com/p/claude-code-extended-thinking): What Effort Levels Cost and What You Can't See
- [Claude Code System Prompt: What's Inside (Token Map)](https://www.claudecodecamp.com/p/inside-claude-code-s-system-prompt): Measured API payload: system text vs tools vs CLAUDE.md, and how to read /context
- [Claude Code Hooks: The Complete Guide](https://www.claudecodecamp.com/p/claude-code-hooks-the-complete-setup-guide): Also, the 5 hooks worth installing today
- [I Can't Stop Running Claude Code Sessions](https://www.claudecodecamp.com/p/i-take-my-laptop-to-the-gym-so-claude-doesn-t-have-downtime): Confessions of an AI coding addict
- [Claude Context Window Explained: 200K vs 1M in Claude Code](https://www.claudecodecamp.com/p/claude-code-1m-context-window): Needle experiments, latency, and when 1M still wins after flat pricing
- [I'm Building Agents That Run While I Sleep](https://www.claudecodecamp.com/p/i-m-building-agents-that-run-while-i-sleep):  I Have No Idea If What They Ship Is Any Good
- [Claude Code MCP: Setup, Token Cost, and When to Skip It](https://www.claudecodecamp.com/p/claude-code-mcp-deep-dive): How Model Context Protocol works in Claude Code, what Tool Search changed, and when Bash is enough
- [How tool use actually works in Claude Code](https://www.claudecodecamp.com/p/how-tool-use-actually-works-in-claude-code): What It Is, Why It Beat RAG for Code, and Why Claude Code Is Just a While Loop
- [Prompt caching notes + remote control + auth (newsletter)](https://www.claudecodecamp.com/p/deep-dive-on-prompt-caching): Digest: caching notes, remote control from your phone, and auth changes. Deep dive linked.
- [How Prompt Caching Actually Works in Claude Code](https://www.claudecodecamp.com/p/how-prompt-caching-actually-works-in-claude-code): The Hidden System That Makes Claude Code 80% Cheaper
- [Agent teams notes (newsletter)](https://www.claudecodecamp.com/p/how-agent-teams-actually-work): Newsletter digest. For the canonical deep dive, see Agent Teams under the hood.
- [Claude Code Agent Teams: How They Work Under the Hood](https://www.claudecodecamp.com/p/claude-code-agent-teams-how-they-work-under-the-hood): Claude Code agent teams vs subagents, the filesystem protocol, cost, and how to debug a stuck team
- [All about Opus 4.6](https://www.claudecodecamp.com/p/all-about-opus-4-6): Opus 4.6 shipped. Agent teams shipped. Your token budget did not survive.
- [How the Claude Code team really works](https://www.claudecodecamp.com/p/how-the-claude-code-team-really-works): Tips from Boris Cherny, the creator of Claude Code
- [Clawdbot broke the internet](https://www.claudecodecamp.com/p/clawdbot-broke-the-internet): Then Anthropic made them change the name
- [Why Jesse  doesn't read code anymore](https://www.claudecodecamp.com/p/why-jesse-doesn-t-read-code-anymore): What I learned from the creator of Superpowers