Keelen
An autonomous coding loop that ships tested pull requests
5 followers
An autonomous coding loop that ships tested pull requests
5 followers
An autonomous coding loop: it plans tasks from your roadmap, writes code in isolated VMs on your own Claude, Codex, GLM, or Kimi key, and ships tested PRs. Every change passes five verification gates before merge: a red-first test proof, an independent adversarial review, the project's own test suite, CI green, and a review window. Flat tiers from $29/mo. Zero token markup.
Hi Product Hunt. I'm Jamie, and I built Keelen.
Most AI coding tools sit in your editor and wait for a prompt. Keelen works from
a roadmap instead. You write a request in plain language, it produces prioritized
roadmap items and dev-ready tasks with acceptance criteria, then implements them
against your GitHub repo around the clock.
The part I care most about is what happens before anything merges. Every change
clears five gates: a red-first proof, where new tests must fail without the
implementation and pass with it (tests born green are rejected); an independent
adversarial review by a model that shares no context with the run that wrote the
code; your own test suite on a clean checkout; CI green at merge time; and a
review window. The loop is never allowed to weaken a check to get to green.
It runs on your own Claude, Codex, GLM, or Kimi key, so tokens bill to your
provider at cost with no markup from us.
You keep the controls. Plan review before any code is written, manual merge, or
branch-only where it never opens a PR at all. Pause anything anytime.
Keelen's own repository runs on Keelen, through the same gates.
Happy to answer anything, including where it falls short.