# claude.dev

> Technical writing from Anthropic for people building with claude:
> deep dives, mental models, and guides.

## posts

- [Claude Code in the cloud: a field guide to cloud sessions](https://claude.dev/blog/claude-code-in-the-cloud/): 2026-10-06 · playbooks · 18 min · [markdown](https://claude.dev/blog/claude-code-in-the-cloud.md) — What changes when Claude Code runs on its own machine, the workflows where that pays off, and how to connect GitHub on the first try.
- [Getting started with Claude Code mods](https://claude.dev/blog/getting-started-with-claude-code-mods/): 2026-10-01 · tutorials · 11 min · [markdown](https://claude.dev/blog/getting-started-with-claude-code-mods.md) — Build your first Claude Code mod from an empty folder, then see what else the API can do.
- [Building with Claude Sonnet 5.5](https://claude.dev/blog/building-with-claude-sonnet-5-5/): 2026-09-28 · playbooks · 9 min · [markdown](https://claude.dev/blog/building-with-claude-sonnet-5-5.md) — When to choose Sonnet over Opus, what it costs, and how to tune it.
- [Automating eval design and hillclimbing with Claude](https://claude.dev/blog/automating-eval-design-and-hillclimbing/): 2026-09-28 · playbooks · 12 min · [markdown](https://claude.dev/blog/automating-eval-design-and-hillclimbing.md) — Principles for designing evals and hillclimbing against them without fooling yourself, and how the claude-api skill's build-eval and hillclimb commands put them to work.
- [What a task costs on Opus 5.5](https://claude.dev/blog/what-a-task-costs-on-opus-5-5/): 2026-09-25 · playbooks · 21 min · [markdown](https://claude.dev/blog/what-a-task-costs-on-opus-5-5.md) — Opus 5.5 costs less per token than Opus 5.
- [Using Claude Code: Spending your effort](https://claude.dev/blog/spending-your-effort/): 2026-09-25 · engineering · 8 min · [markdown](https://claude.dev/blog/spending-your-effort.md) — What effort really is and when to use which level in Claude Code, from my own tests of three builds and a deep dive into Terminal-Bench 3.0 on Opus 5.5 and Fable 5.1.
- [How we made claude.ai 3x faster in two weeks](https://claude.dev/blog/how-we-made-claude-ai-faster/): 2026-09-23 · engineering · 15 min · [markdown](https://claude.dev/blog/how-we-made-claude-ai-faster.md) — Once Claude can measure something, it can make it faster. So we kept finding more things to measure.
- [Getting the most out of Opus 5.5 in Claude and Claude Code](https://claude.dev/blog/getting-the-most-out-of-opus-5-5/): 2026-09-22 · playbooks · 9 min · [markdown](https://claude.dev/blog/getting-the-most-out-of-opus-5-5.md) — How to prompt Opus 5.5, steer a long run, and check your results in Claude apps and Claude Code.
- [The new rules of context engineering for Claude 5 generation models](https://claude.dev/blog/the-new-rules-of-context-engineering-for-claude-5-generation-models/): 2026-07-24 · engineering · 7 min · [markdown](https://claude.dev/blog/the-new-rules-of-context-engineering-for-claude-5-generation-models.md) — We removed over 80% of Claude Code's system prompt for more advanced models. How to apply the lessons we learned to your own context engineering in Claude Code and with your own agents.
- [Lessons from building Claude Code: How we use skills](https://claude.dev/blog/lessons-from-building-claude-code-how-we-use-skills/): 2026-06-03 · skills · 11 min · [markdown](https://claude.dev/blog/lessons-from-building-claude-code-how-we-use-skills.md) — What we learned building and scaling hundreds of skills internally at Anthropic.
- [A harness for every task: dynamic workflows in Claude Code](https://claude.dev/blog/a-harness-for-every-task-dynamic-workflows-in-claude-code/): 2026-06-02 · agents · 11 min · [markdown](https://claude.dev/blog/a-harness-for-every-task-dynamic-workflows-in-claude-code.md) — Claude Code can now write and orchestrate its own multi-agent harness on the fly. Here's how dynamic workflows work, and the patterns that get the most out of them.
- [Using Claude Code: The unreasonable effectiveness of HTML](https://claude.dev/blog/using-claude-code-the-unreasonable-effectiveness-of-html/): 2026-05-20 · playbooks · 9 min · [markdown](https://claude.dev/blog/using-claude-code-the-unreasonable-effectiveness-of-html.md) — How and why members of the Claude Code team use HTML instead of Markdown to produce richer, more readable, and easily shareable outputs.
- [Lessons from building Claude Code: Prompt caching is everything](https://claude.dev/blog/lessons-from-building-claude-code-prompt-caching-is-everything/): 2026-04-30 · engineering · 6 min · [markdown](https://claude.dev/blog/lessons-from-building-claude-code-prompt-caching-is-everything.md) — Best practices for optimizing prompt caching in Claude Code, including how to most effectively structure your prompt, use tools, and layer on compaction.
- [Seeing like an agent: how we design tools in Claude Code](https://claude.dev/blog/seeing-like-an-agent/): 2026-04-10 · agents · 7 min · [markdown](https://claude.dev/blog/seeing-like-an-agent.md) — Learn how the Claude Code team designs, tests, and evolves tools by thinking from the model's point of view.

## other

- [posts.json](https://claude.dev/posts.json): the post index as data
- [rss.xml](https://claude.dev/rss.xml): the published posts as an RSS 2.0 feed