Claude Code
Terminal coding agent that works in a real checkout.
- made by
- Anthropic
Magic Ship is one shop in Vancouver, BC, working remotely with clients worldwide. We are not a partner, reseller, or certified vendor of Anthropic - we just build with this.
What Claude Code is
Claude Code is Anthropic's coding agent. It runs in a terminal or IDE against a real working copy, reads and edits files, runs commands, and takes standing instructions from CLAUDE.md files checked into the repo. Skills, subagents, hooks, and MCP servers extend it per project.
How we use it
Most of our codebase work is aimed at this class of tool: we write the repo's instruction files, scope rules to the paths they apply to, and make sure the checks an agent needs - typecheck, lint, tests, a one-command local stack - run without a human in the way. Teams get worked examples in their own codebase rather than a slide deck. Then we look at whether the generated changes survive review, because that is the only number that matters.
Where it is the wrong choice
It is only as good as the feedback loop around it. In a repo where the tests take forty minutes or cannot run locally at all, it produces confident changes nobody can verify, and the honest first engagement is fixing the loop rather than installing the agent.
Service lines it turns up in
Related tools
More in Agent frameworks
Other tools in the same service lines
Building something on Claude Code?
Send the problem rather than a job spec. You get an answer on scope, on fit, and on whetherClaude Code is even the right call for it.
Start a projectClaude Code and Anthropic are trademarks of their respective owners, used here to say what we work with.