Queue the task. Agents take it.
Queue the task. Agents do the work.
PREQSTATION is a self-hosted execution system for Claude, Codex, and Gemini. It is not a generic issue tracker: scoped tasks move through inbox, todo, ready, and done, while hold branches off blocked work and queued or working stay separate as run-state overlays.
Execution snapshot
PREQ runs a happy path from inbox to done, then branches blocked work into hold instead of inventing another fake in-progress column.
Inbox
inbox
Todo
todo
Ready
ready
Done
done
Hold
blocked branch
Move work here only when implementation or verification is blocked. Once the blocker is cleared, it can rejoin the path to ready.
Open Source · Self-hosted · Multi AI Agent
#3953
Rate Limit API
#3954
Auth Refactor
#3955
Deploy Script
The missing part is execution discipline
AI can write code. Shipping still breaks down between the idea and the verified result.
Issue trackers stop at assignment
They hold backlog and discussion, but they do not execute the task.
Raw AI sessions lack workflow guardrails
No queue, no branch policy, no review state, no audit trail.
Simple work still needs a safe workspace
Agents need isolated worktrees, repeatable steps, and reporting back.
The handoff gap burns the schedule
The delay is not writing code. It is routing, checking, and deciding what can ship.
Work like a kitchen.You are the chef.
Who is it for?
Ideas on the go
Fire AI tasks from anywhere, anytime on mobile
Multi-project juggler
Vibe code across projects without context loss
Context switcher
Freelancing? No more 'What did I do here?'
Multi-engine strategist
Pick Claude, Codex, or Gemini per task
Quality-focused dev
Systematically review and verify AI output
Need simple issue tracker
Use GitHub Issues, Linear, or Jira for raw backlog management. PREQ assumes the task is already scoped.
Want team collab tool
Use Notion or Asana first. PREQ is for execution, not general collaboration.
Don't use AI for coding
Preq is built for AI-driven workflows
Single project, slow pace
Works best with ongoing, dynamic work
Need sprint management
PREQ executes tasks; it does not replace roadmap or sprint planning.
Built for developers who ship
Use GitHub Issues, Linear, or Jira for backlog management. PREQ starts when the task is scoped enough to execute reliably.
Execution instead of issue grooming
PREQ is where a scoped task becomes runnable work with claim, branch, result, and review attached.
Pick the right engine per task
Assign Claude, Codex, or Gemini per job instead of forcing one model into every task shape.
Ship from isolated worktrees
Agents do not write directly into your main branch. They work in controlled sandboxes and report back.
Keep an audit trail before merge
Work logs, tests, and branch details make review concrete. You know what happened before something reaches done.
Why engineers actually use it
PREQ turns a scoped task into a repeatable execution loop.
Turn scoped work into runnable tasks
Push a cleaned-up request into inbox from the web app or Telegram once the task is worth executing.
Give agents a controlled workspace
Agents claim work with run_state, use isolated worktrees, and return notes, tests, and branch output instead of hand-wavy status updates.
Treat hold as a real branch
Blocked work leaves the happy path, waits in hold, and returns only after the blocker is actually resolved.
Review before done
Ready means implementation submitted, not trusted. You still verify checks, inspect the diff, and decide what ships.
Execution Workflow
Happy path goes inbox → todo → ready → done. Hold branches off when work is blocked.
Inbox
inbox
Todo
todo
Ready
ready
Done
done
Hold
blocked branch
Tasks move here when implementation or verification is blocked. They rejoin the happy path only after the blocker is cleared.
Run-state overlay
Use queued when dispatch is requested and working once an agent claims the task. They sit on top of status instead of creating fake workflow columns.
Not an issue tracker
Use GitHub Issues, Linear, or Jira to discover and triage work. PREQ starts once the task is scoped enough to execute safely.
Choose the right AI for the task
Pick the best tool, no vendor lock-in
Strong at complex refactoring, architecture changes, and general tasks
Strong at code review, security analysis, and architecture review
Strong at UI/UX design, documentation, and large-context processing
Free. Open Source. Forever.
Your code, your data, your server
- ✓Full source code on GitHub
- ✓Deploy on your own server
- ✓Community-driven standards
- ✓Your data stays yours
- ✓No vendor lock-in
Frequently Asked Questions
Open your kitchen today
Read the guide, connect your agents, and turn scoped tasks into verified execution.