Claude Code is Anthropic’s agentic coding tool. It runs in the terminal, reads the codebase, edits files, and runs commands, and the same engine powers IDE extensions for VS Code and JetBrains, a desktop app, a web surface, GitHub Actions, and Slack, with session handoff between them.
models and pricing
Claude Code runs Claude models and defaults to Opus 5.5 on Anthropic’s own plans. It is included in the paid individual Claude plans, Pro at $20/mo, Max at $100/mo, and Ultra at $200/mo, and in Team seats. It also runs on Anthropic Console API keys, pay per token. Enterprise routes through Amazon Bedrock, Google Cloud, and Microsoft Foundry are documented; non-Claude models, local ones included, are not supported.
what it’s good at
Claude Code leads the category on orchestration surface: subagents with a coordinating lead agent, background agents, lifecycle hooks that run shell commands around its actions, MCP client support, and skills and plugins. Project memory lives in CLAUDE.md and AGENTS.md files, and the Agent SDK lets you build custom agents on the same engine.
Claude Code in crystl
Claude Code is the agent crystl treats as the reference integration. Its hook surface is the one the approval cards were designed around, and its session ids are what let a shard survive a restart with its conversation intact.
Everything the catalog says about it is description. Which binary runs, which hooks are installed, and which model a shard launches on are all resolved on your own machine.