Superplan is an execution system for AI agents.
It turns planning into actual execution inside your repository. Instead of vague plans, chat history, or TODO lists, Superplan forces work into clear, step-by-step tasks that agents can execute, track, and resume at any time.
The CLI is designed for agents to follow, not for humans to run manually. You define the work; your agent executes it through Superplan's structured runtime.
Most work fails because:
- plans drift across chats and notes
- steps get skipped
- progress is unclear
- AI agents lose direction
Superplan fixes this by making work explicit and structured.
curl -fsSL https://raw.githubusercontent.com/superplan-md/superplan-plugin/main/scripts/install.sh | shcurl.exe -fsSL -o install-superplan.cmd https://raw.githubusercontent.com/superplan-md/superplan-plugin/main/scripts/install.cmd; if ($LASTEXITCODE -eq 0) { .\install-superplan.cmd }curl.exe -fsSL -o install-superplan.cmd https://raw.githubusercontent.com/superplan-md/superplan-plugin/main/scripts/install.cmd && install-superplan.cmdAfter install, Superplan will ask if you want to run superplan init.
superplan init
Superplan lives inside your repo and controls execution.
- You define a change
- Superplan converts it into tasks
- Tasks are executed one by one
- Progress is saved and tracked
Everything is stored in:
.superplan/changes/
This means:
- no lost context
- no guessing what's next
- work can be resumed anytime
Create a change:
superplan change new my-feature --jsonDefine tasks inside:
.superplan/changes/my-feature/tasks.md
Run execution:
superplan run --jsonTrack progress:
superplan status --jsonComplete tasks:
superplan task complete <id>
superplan task approve <id>| Normal planning | Superplan |
|---|---|
| Notes scattered everywhere | Tasks stored in repo |
| Next step is guessed | CLI tells you what's next |
| Progress unclear | State is tracked |
| Hard to resume | Fully resumable |
Superplan ensures:
- every step is defined
- every task is tracked
- nothing is skipped
- Developers who want structured execution
- Teams working on complex features
- Anyone using AI agents and tired of drift
Superplan uses internal "skills" to manage execution:
- planning and task creation
- execution loop
- debugging and recovery
- validation and review
You don't need to learn these to get started, but they enforce structure behind the scenes.
Superplan helps you stop guessing what to do next and actually finish work.
Apache License 2.0. See LICENSE for details.