MindFlock

About

I built MindFlock because I kept losing track of my own AI agents. I usually have three or four going at once. The bottleneck was never the models — it was me. Six terminal windows, no idea which one needed my attention, and every session starting with the same ritual: pick a ticket, make a worktree, write a prompt, wait. MindFlock polls your tracker — Jira, Linear, GitHub Issues, Shortcut, Asana — for work assigned to you, and turns each ticket into a real session: its own git worktree on its own branch, dependencies installed, an agent already seeded with the ticket's description, acceptance criteria, and comments. Nothing typed. PRs that come back with review comments become sessions the same way. You come in where a human is needed: read the diff, then commit → push → PR → merge, one click each. Two things I got wrong that shaped the design: Running multiple agents in one repo breaks confusingly — two sessions edit the same file, one reads a stale version of what the other just wrote, and it looks like the model being dumb when it's your setup. Worktrees fixed it properly. Reading terminal output to detect agent state is unreliable — it misses anything that isn't a visible prompt string. The agent CLIs' own lifecycle hooks made state deterministic, and show what screen-scraping can't, like "commit hooks are running right now." The decision I'd defend hardest: MindFlock never commits, pushes, opens or merges by itself, and never writes to your tracker. Every one of those is your click. A pipeline you can't fully trust is worse than no pipeline. It also runs entirely on your machine — no cloud sandbox, no code uploaded, your own agent subscription. Apache-2.0, free, no accounts. Built solo on evenings and weekends, used daily for a year. I'd love feedback, especially from anyone running agents in parallel. Where did you hit the ceiling? Three or four is where I stop reviewing as fast as they produce — is that universal or just me?

Problem & solution

Who it's for

Developers already running multiple AI coding agents who review every diff themselves.

Share your thoughts about this tool.

No comments yet. Be the first to leave one.