If you use Linear and Claude Code or Codex, you've probably wished you could hand off a ticket to an AI agent the same way you'd assign it to a teammate: open the issue, click assign, done.
That's what Delego does.
The problem
AI coding agents exist in two forms today, and both have real limitations.
Cloud agents (Codex Cloud, Claude.ai) run in isolated environments. They can't reach your private repos without explicit grants, can't touch services on your internal network, and can't use your local SSH keys or shell environment. They also constrain your model choices — the platform decides which model runs your job, not you. And there's account risk: heavy automated use through a cloud platform can look like unusual activity to the provider.
Running agents locally by hand is the alternative, but it doesn't scale. You trigger the run, watch the terminal, and stay in the loop manually. You switch between Linear (to understand the task) and your terminal (to run the agent), then come back to Linear to post an update. Context-switching is the whole job.
What Delego does
Delego bridges the gap. You label a Linear issue, and a paired runner on your own machine picks it up, executes Claude Code or Codex in an isolated git worktree, and reports progress back to the Linear Agent Session — all without you leaving the issue view.
The flow looks like this:
- Open a Linear issue. Add a label to delegate it —
executor › claudeorexecutor › codex, plus an optional model label likemodel › sonnet. - A paired runner on your machine polls for the job and claims it.
- The agent runs in a fresh git worktree. Your current branch and uncommitted changes are never touched.
- Progress appears as comments on the Linear issue in real time.
- When it's done, you get a pull request to review.
Start to finish, you stay inside Linear. There's no second tool to check, no separate inbox, no terminal to babysit.
Why it's better than running cloud agents
Your full environment. Cloud agents run in sandboxes. A Delego runner inherits your entire shell environment. If it works in your terminal — internal services, private repos, local SSH keys — it works through Delego.
Your model choices. Cloud agent platforms pick the model for you. With Delego, model selection is just a Linear label you set per job. Run opus on a complex refactor and sonnet on the quick fix. Change it issue by issue without touching any configuration.
No account risk. Delego shells out to your own claude or codex CLI using your own logged-in session — exactly the same as running it by hand in a terminal. The provider sees normal terminal usage, because that's exactly what it is.
No vendor lock-in. You can run Claude Code on one issue and Codex on the next. Whichever agent fits the task, whenever you want.
Parallel execution. Pair multiple runners — on the same machine or across machines — and they pull from the same queue simultaneously. More runners means more jobs running in parallel.
Why Linear
We built Delego on top of Linear instead of building a standalone task tool for a few reasons.
Linear is already mature. It handles the hard parts of issue tracking — keyboard-first UX, projects, cycles, triage, mobile — and your team already knows how to use it. You don't have to introduce another tool.
Linear is a knowledge base. Issues carry rich context: linked docs, related issues, comments, attachments, discussion threads. When an agent picks up a ticket, it inherits all of that context automatically. The work the agent does stays attached to the issue permanently, searchable alongside everything else.
Linear is investing in agentic development. Agent Sessions, the webhook API, the integration model — these are first-class Linear features, not bolted on. Building on top of that means Delego benefits from Linear's product direction instead of fighting it.
Getting started
You need four things:
- A Linear workspace (the free tier supports up to 250 active issues — more than enough to evaluate Delego on a real project)
- Claude Code or Codex installed and working locally
- Node 20+ on the machine you want to run jobs on
- A Delego account, paired to a runner
The installation guide walks through each step. Total setup time is under ten minutes.
If you've been running agents manually from the terminal, or you've been looking for a way to put Linear at the center of your agentic workflow without cloud middlemen, Delego is worth trying.