Run background coding agents on the subscription you already pay for

Updated 2026-10-23

Denly uses the supported authentication configured for a runtime CLI on your runner machine. AI usage stays with that provider, subject to its plan or API billing. Subscription access and cloud execution can coexist; where tools run is a separate choice.

Denly is a control plane for coding agents (Claude Code, Codex, Cursor, Grok Build, Antigravity) that run on your own machines using your vendor subscription or API key. It manages tasks, plans and reviews across projects; Git-backed tasks can open pull requests, while no-Git projects end with a review of local changes. Inference is billed directly by your runtime provider; Denly charges for the control plane and nothing else.

Subscription access is separate from execution location

A first-party service can include both local and hosted agent usage in its own subscription. For example, OpenAI's pricing and usage documentation describes shared allowances for local messages and cloud chats. Eligibility, limits and additional charges depend on the plan; API-key usage has separate billing.

A third-party service has its own supported authentication and billing arrangements. Do not assume your existing subscription works there, or that it must require an API key: check that service and your provider's terms.

Denly invokes the CLI on your runner machine using its supported login or API-key configuration. It does not transfer that login to a hosted agent or extend your provider's allowance. Check denly authmode and your provider's usage dashboard before starting unattended work.

The OpenAI billing description was checked against the official documentation linked above on 23 October 2026. Check the provider's current terms and limits before deciding.

The five runtimes, and what each reports

Usage shows the quota a provider reports plus the tokens Denly counted, separated into task work and Chat. Two of the five publish a quota; the other three do not, and Denly shows what it has rather than an estimate.

Set the runtime, per project or per workspace

1. Install and authenticate the runtime CLI on the runner machine. denly authmode reports how each installed CLI is authenticated there.
2. Pick the workspace default runtime and model, then override it per project where a project needs a different one.
3. To use an API key instead of a subscription, denly set-key stores it in the machine's own secret store – not in a file, and not on Denly's servers. denly clear-key removes it.
4. A task can also name its runtime at creation, and a failed run can be retried fresh on a different one.

What Denly costs on top

The control plane is what you pay Denly for, and the runtime provider bills your inference separately:

These are the published monthly prices. Check the pricing section and checkout for available billing intervals, current prices and taxes. A Denly plan does not include extra AI-provider usage.

The trade-off

Denly executes work only while the selected runner machine is awake and online. Queued work waits; an interrupted active task may need recovery. Use an available machine you manage or a hosted service when work must continue while your laptop is off. Subscription eligibility, usage limits and execution availability are separate constraints.

If your goal is simply to continue a local Claude Code session from your phone, native Remote Control may be enough. See when to use Remote Control and when Denly adds value before installing another service.

Read next

AI runtimes & models · Security, encryption & data · Run Claude Code on your own machine, and drive it from anywhere · End-to-end encrypted coding agent tasks: what the server can and cannot read

Start free – no card