Overview
Jardinero is an open-source, self-hostable AI engineer that tends a codebase. It is described as a crew of always-on AI engineers that watch logs, reproduce what breaks in disposable sandboxes, and ship fixes. It runs a factory of small, focused machines, each watching a different door into the codebase and shipping pull requests for review.
Jardinero performs three main jobs. It reviews production telemetry from Grafana logs and metrics every few minutes, scoring signals against an investigation threshold and rejecting noise with a reason. It implements assigned Linear issues by scoping the task, building it in an isolated sandbox, running tests, and handing back a review-ready PR. It also maintains GitHub issues and pull requests when tagged @jardinero, answering review threads with evidence, fixing failing checks, resolving conflicts, and following repo conventions.
Every job happens in a disposable sandbox with a pool running in parallel, and nothing touches branches until the work is green. Jardinero is built by Tenki, part of Luxor, and tends Tenki’s own codebase daily using Tenki sandboxes. It is open source under LuxorLabs, can be self-hosted on a laptop or server, and run reports can land in Discord.
Pricing
Free
Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.
Visit WebsitePrices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.
Use Cases
- Teams use Jardinero to review production telemetry from Grafana logs and metrics around the clock.
- Teams use Jardinero to reproduce and fix real production issues in disposable sandboxes.
- Teams use Jardinero to implement assigned Linear issues and receive review-ready pull requests.
- Teams use Jardinero to maintain GitHub issues and pull requests by tagging @jardinero.
- Teams use Jardinero to run jobs in parallel disposable sandboxes without touching branches until green.
- Teams use Jardinero to self-host the AI engineer on their own laptop or server.
- Teams use Jardinero to receive run reports in Discord.
- Teams use Jardinero to queue only exhausted retries or judgment calls for human attention.
Features
AI assistant
Natural language interface
Workflow automation
Reviews
No reviews yet. Be the first to share your experience!