
Keelen· AI coding agent for GitHub
AI coding agent for GitHub, steered from chat through MCP
Developer Tools·AI & Machine Learning
What is Keelen?
Keelen is an AI coding agent for GitHub.
Keelen is an AI coding agent for GitHub, steered from chat through a hosted MCP server. Turn requests into a standing roadmap, choose plan approval and merge controls, and review pull requests. Connect your own supported AI subscription or API key with no markup on AI usage.
Connect an MCP client to using Streamable HTTP. Public setup documentation is at , with guides for clients including Claude Code, Claude Desktop, Cursor, Windsurf, and the Codex CLI. The two onboarding tools, signup and verify_email, work without a key. Your assistant can guide signup, while email verification, AI credential connection, and GitHub approval are completed in the browser. Workspace tools require an authenticated account.
A request submitted through MCP enters the same intake path as the Keelen dashboard. The planner turns it into tasks with acceptance criteria. Development runs work in single-use virtual machines, and another model reviews the diff. Required repository checks and the connected project’s merge settings decide when work can land. You can choose approval controls, check roadmap progress, answer questions, review pull requests, and pause or resume work.
Keelen is for people with a GitHub repository who want to keep a backlog moving while steering requests and decisions from chat. It supports bringing your own AI subscription or API key using the methods each provider supports. The hosted backend is proprietary; the public keelen-mcp repository contains setup documentation, not the backend source. Connecting GitHub requires explicit GitHub App approval, and keys can be revoked in the dashboard. See for the setup flow and for current pricing.
Similar to Keelen
More Developer Tools toolsJoryio
Marketing automation, evolved for the AI era
Nick Launches Directory Finder
156 free launch directories, sorted by domain rating
Plutou
The AI-native operations platform for collection operators.
Modeinspect
The AI-native way to design production-grade UI.
Analyse
Analytics, an AI SEO engine, and a copilot in one subscription
specgit
Edit GitHub docs without learning Git or Markdown
Ready to launch?
Submit your product and get discovered by builders and creators worldwide.
Launch NowFounderPlaybooks.
What other founders did to grow.
2722 dispatches from hundreds of founders, pulled from the week's best podcasts.
I realized that I could make it even easier for people. So I started building these no-code integrations — for even less technical people to use Data Fetcher. And that drove it through to 10,000 in MRR.
No-code integrations took MRR from 3K to 10K by unlocking less-technical users
Removing technical setup friction expanded the activatable audience beyond developers. Pre-built no-code integrations took MRR from 3K to 10K by letting less technical users reach value in their first session — same product, broader funnel.
Catching something early on could be 1x of your cost and then 10x if you're fixing it by clean up downstream — because now you've got new data coming in that you'd have to continually add new iterative processes to fix — and then lastly at the last mile maybe 100x more cost if you have bad reporting that are actually telling the wrong things.
Fix Data Architecture Early: Upstream 1x Cost, Downstream 100x Cost
Bad data compounds like debt. Taylor Wells worked on Disney+ from the ground up and saw this pattern at enterprise scale: a misnamed event or wrong tracking structure that costs an hour to fix at launch costs ten hours mid-product and a hundred hours — plus wrong business decisions — at the reporting layer. The principle applies to solo indie developers: spend half a day thinking through your event taxonomy before you ship.
There's a play for whatever you're stuck on.
Read all 2722 playbooks
Comments
0