GuideSeptember 18, 2026

Using Claude Code, Codex, and OpenCode with Denker

Denker provides the workspace, your AI CLI provides the intelligence.

Denker is a new interface for AI agents: instead of typing into a chat window, you point, draw, or talk on your screen, and Denker's agents see what you mean. They work across everyday apps like Gmail, Figma, and Sheets and bring the result back where you are. Coding work can run on your own Claude Code, Codex, or OpenCode. This guide shows what Denker adds in front of Claude Code, Codex, and OpenCode, how to mix them per task, and what each part costs.

INTERFACE

What Denker adds as the interface layer

Denker is the new interface for AI agents. You hand off work by voice or text from the app where you are, Denker sees what you mean, and your CLI does the work across everyday apps like Gmail, Figma, and Sheets — the result comes back where you are. The flow is Point. Dispatch. Get it back.

Denker handles where and how work runs, your CLI handles the thinking.

CLAUDE CODE

Claude Code with Denker

With Claude Code, Denker acts as the interface layer in front of the subscription you already pay for. Point at work on your screen, dispatch by voice or text, and Denker gives Claude Code the on-screen context plus the ability to act across your apps. For example, in one demo a user circles empty space on a site preview, and Denker edits Hero.tsx and highlights the new Book a consultation button.

Keep Claude Code, add on-screen handoff and cross-app results.

CODEX

Codex with Denker

With Codex, the pattern is the same. Stay in the app where you already work, point or draw to show what you mean, and talk or type the handoff. Denker dispatches Codex with that screen context and brings the result back where you are working. For example, with three competitor pricing pages open, Denker returns a report with a price table and a team plan chart.

Codex intelligence, dispatched from the app where you already work.

OPENCODE

OpenCode with Denker

With OpenCode, Denker adds on-screen handoff and cross-app follow-through to the setup you already use. Hand off work by voice or text, pointing at items on screen or circling them to show what you mean. For example, a user dispatches three agents, then watches a login redirect fix move to Done while its pull request checks pass.

OpenCode power with point, draw, and talk context.

CHOICE

Pick per task and work side by side

You do not have to pick just one CLI forever. Pick one per task, or run them side by side and follow their progress on your screen — one agent on Figma guidance while another drafts a Gmail reply and a third helps with a Sheet. Denker routes each handoff to the CLI you chose.

Mix and match CLIs without changing where you work.

PRICING

Subscription and cost honesty

Cost has two parts, kept separate:

Denker: Free €0 (25 on-screen calls, 2 projects, 5 tools) or Pro €19/month (unlimited), with a 14-day trial and no credit card.
Your provider: the Claude Code, Codex, or OpenCode subscriptions you already pay, billed by them — Denker never resells tokens or bundles model fees.

Pay Denker for workspace, pay your provider for intelligence.

This guide supersedes and expands the thin multi-CLI support post with full setup, per-CLI guidance, and cost detail.

Frequently asked questions

Which AI CLIs does Denker support?

Denker works with Claude Code, Codex, and OpenCode. Pick one per task, or run them side by side and follow their progress on your screen.

Do I need an API key?

No. Denker uses the subscription and login you already have for your coding CLI, so there is nothing new to set up.

What does Denker cost, and what about model costs?

Free is €0 with 25 on-screen calls per month, 2 projects, and 5 connected tools. Pro is €19 per month with unlimited calls, with a 14-day Pro trial and no credit card. AI provider charges are separate, and the included AI starter allowance is limited.

Where does work happen and how do I stay in control?

Denker works on your screen and across your apps: point at what you mean, and the result comes back where you are working. Denker asks before it moves your pointer, shows its cursor while it works, and never sends messages or emails on its own. Drafts wait for you to review and send.