--- title: "Auteur works with Claude Code" canonical_url: https://auteurintelligence.com/integrations/claude-code/ description: "Connect Claude Code to Auteur and keep working the way you do now, with persistent projects, governed campaigns, verification, and receipts underneath. The…" --- # Auteur works with Claude Code Connect Claude Code to Auteur and keep working the way you do now, with persistent projects, governed campaigns, verification, and receipts underneath. The agent can change; the project, process, authority, and evidence remain. --- ## What Claude Code does well Claude Code is a strong interactive coding agent with first-class MCP support and a headless mode that emits a machine-readable stream. It is comfortable running long, multi-step engineering work, which is precisely why Auteur both connects to it as a client and drives it as a campaign seat. --- ## What it brings to the seat Claude Code seats as an orchestrator with real judgment: in our own build ledger it has run hundreds of governed work orders as planner, integrator, and cross-family reviewer. Bring it your way: sign in with your Claude subscription through the CLI, or connect an API key; extra usage rides your own plan, never a middleman meter. Its long-context reasoning makes it the natural conductor for multi-round campaigns. “Claude, how much did this build cost in time and money?” You, at the end of a build night “385k raw tokens, with 3.5 million more served from cache. Wall clock was 3 hours 12 minutes, almost half the DeepSeek V4 time, though V4 finished its round for $3.24 while Grok 4.6 cost $5.12. Gemini Flash 3.7 closed its round in 2 minutes 10 seconds but dropped the tail again, so I seated Opus 4.8 by API as auditor. Eval and memory are updated for the next build.” Illustrative exchange. The receipts, cross-model comparison, and reseating behavior are shipped surfaces. “Claude, you cannot generate video. Give me a ten second establishing shot of the ferry at dawn.” The ask a chat window cannot serve “Alone, no. Seated here, I routed it to the video lane on the providers you hold, staged three candidates on the Visual Canvas, and visual eval scored the takes. Two passed. They are waiting at your gate.” Illustrative exchange. Video routing, canvas staging, and visual eval are shipped surfaces. ## What Auteur adds For someone already using Claude Code, Auteur adds a persistent project that keeps its own canon and evidence instead of losing them to the session; governed multi-round campaigns with acceptance gates and recovery; production tools beyond code; and canonical receipts that verify what actually changed independently of the agent's own claim. Claude Code stays your interface, Auteur is the amplification and accountability layer behind it. --- ## How it connects - **Client.** Claude Code operates authorized Auteur domains as an MCP client. - **Executor.** Auteur drives Claude Code as a Build campaign seat under an executor profile, headless. - **Direct model.** Auteur can call the Anthropic API directly for measured, accounted work, with no harness in the loop. --- ## Which Auteur products it reaches As a client, Claude Code can address the Auteur domains you are entitled to, writing and story, prep, film, game, learning, and the Build campaign system, in natural language. As an executor it runs Build campaign seats. Reachability is not entitlement: the connection can address these domains; which you operate depends on what you hold. --- ## What is measurable When Claude Code runs as an Auteur Build executor or through the Anthropic direct-model lane, Auteur derives token and cost accounting from the record of the work it ran. What you do in Claude Code outside an Auteur campaign is not visible to Auteur. Auteur never invents missing usage data; missing = unknown, never zero. No token or cost figure is ever shown for work performed outside Auteur's visibility. --- ## What this integration does not do - It does not account for Claude Code sessions you run outside Auteur. --- ## Security and authority boundary Connecting Claude Code does not grant it wildcard authority over your files, spending, publication, or application changes. The connection is an interface, not a grant of authority, what a connected agent may do stays governed by Auteur, not widened by connecting it. --- ## Current status Current per-mode status is shown above. --- ## Proof No published Claude Code compatibility canary yet. A recorded, Claude Code-specific evidence artifact will be linked here when one exists, follow the [roadmap](https://auteurintelligence.com/roadmap/) for progress. --- ## Set it up To operate Auteur from Claude Code as a client, the invited local Conductor connection is the supported route: **Request Invitation**, [/founding-beta/?track=invite](https://auteurintelligence.com/founding-beta/?track=invite). A small, deliberately limited cohort; this starts a conversation with us, and seats, when offered, are paid. To run Claude Code as a Build executor inside a campaign, start with **Apply for the Founding Beta**, [/founding-beta/](https://auteurintelligence.com/founding-beta/). --- ## Affiliation Auteur works with, and is designed to connect with, Claude Code. No partnership, endorsement, or certification is claimed, and none exists.