#claude-code
Anthropic’s coding agent: it reads your project and writes or changes code from instructions in plain language. Full definition →
Claude Code models, effort and fast mode: which knob to turn, and when
The model sets how capable an answer can be; effort sets how much work goes into it. Before you turn either, work out why the answer failed. Then: the model aliases, which effort setting overrides which, and what /fast costs if you switch it on at the wrong moment.
Claude Code in a dev container: three layers, and what they do not protect
A dev container holds the whole Claude Code process, not just its shell commands. Add the CLI with one feature line, keep your login across rebuilds with a named volume, and save runs without prompts for the reference container with a firewall and a regular user.
CLAUDE.md that scales: @imports, per-folder rules and what actually loads
In a growing project, CLAUDE.md swells and every line loads into every session. @path imports keep it organised, rules in .claude/rules with a paths field load only when Claude reads a matching file, and /context all shows what really made it in.
Claude Code hooks: PreToolUse guards and three recipes that pay off
A rule in CLAUDE.md is advice Claude can miss. A hook is your own command that Claude Code runs at a set moment, every time. How a PreToolUse hook blocks an edit with exit code 2, three recipes worth adding today, and the boundary people forget.
Subagents, workflows or agent teams: how to choose in Claude Code
Claude Code has three ways to split work between workers: subagents, dynamic workflows and agent teams. The real difference is who holds the plan. Three questions, asked in order, tell you which one a task needs — and every step up costs more tokens.
Claude Code worktrees: run parallel sessions without collisions
Two Claude Code sessions in one folder edit the same files and trip over each other. claude --worktree gives each session its own working copy on its own branch. Here is how to prepare a project, the trap with unpushed commits, and how to bring the work back.
Claude Code 2.1.311: what's new
Claude Code 2.1.311 improves startup speed, session handling, and accessibility, with better shell command feedback and model effort adjustments.
Claude Code 2.1.308: what's new
Claude Code 2.1.308 improves API versioning, subagent coordination, and recovery from errors, with better settings control and sign-in experience.
The Claude Code practicum on GitHub: 14 hands-on tasks with automatic checks
A free, open practicum for the Claude Code: Pro Level course: 14 tasks on a small real project that you do in your own terminal with the real Claude Code, while npm run check tells you what passed. Finish all 14 and you get a code for the Claude Code Practitioner badge.
Claude Code 2.1.304: what's new
Claude Code 2.1.304 brings new warnings for usage limits, changes to skill hooks and compaction, and removes the "(research preview)" label from `/fast` on Opus 5.5.
Claude Code artifacts turn AI sessions into shareable web pages
Claude Code can now package your AI coding sessions as live web pages that update automatically, making collaboration easier.