Hi Product Hunt! 👋 I'm Odilkhon, founder of Quiel. We built it for a moment a lot of teams are hitting right now: everyone on the project runs their own coding agent, and suddenly nobody knows who is doing what. Two agents pick up the same task. A risky command runs that nobody approved. And when something breaks, there is no record of what ran, where, or why. Quiel is one task queue for your team and its agents: • Bring the agents you already use. Claude Code, Codex, Cursor and OpenCode connect through one client, @quiel/cli. Nobody has to switch. • Nothing runs twice. One dispatcher hands each task to exactly one agent with a lease. If a laptop goes to sleep, the task goes back to the queue. • A person decides what is risky. Stop patterns and trust profiles pause dangerous commands until a named teammate approves, from the web app, Telegram or Slack. • Your code stays on your machines. Agents run locally on your own subscriptions; Quiel only sees file names, paths and commands. • Every decision is on the record, in a journal nobody can edit or delete. You can also import a spec and get a task graph with roles and dependencies in one step, and connect GitHub or GitLab for branches, pull requests and checks. Pricing is per team, not per seat, and there is no usage meter. Your agents run on your own subscriptions, so we have no inference cost to pass on. The Free plan covers up to 3 people. Setup takes about 15 minutes: sign up at quiel.app, then run npm i -g @quiel/cli and quiel init in your repo. We'd love your honest feedback, especially this: which commands would you never let an agent run without asking you first? Thanks for checking us out! 🙏

Quiel

Run your team’s coding agents on one queue, humans approve

Run your team’s coding agents on one queue, humans approve