Project Workflow · Lifecycle control

ProjectPilot

Track project phases, required artefacts and human approval gates without allowing the tool to execute agents.

Source and README ↗Technical reference ↗

PythonLifecycleLocal-first

Why it matters

The problem

It is easy to lose track of a project's current phase, outstanding evidence and human approvals across different coding assistants.

How it works

The approach

A local deterministic process conductor records lifecycle state, checks phase requirements, recommends next actions and prepares prompts.

What it includes

Explicit idea-to-done lifecycle
Phase requirements and evidence inventory
Deterministic next-step advisor
Readable local dashboard and JSON output

Reproducible workflow · documented commands

What happens after you create a project?

The following is a documented walkthrough, not a screenshot or a transcript from a newly executed CLI run. The project's README and workflow guide explain each step.

1. Initialize

Run pp init to create local lifecycle state for a new project.

2. Inspect

pp status and pp next show the current phase and the recommended next action.

3. Check evidence

pp phase check lists expected artefacts; pp artifact add records evidence metadata after human-created work.

pp init "a local code review assistant" --name "Review Assistant"
pp status
pp next
pp dashboard

Commands are taken from the project README. To reproduce this walkthrough, follow the install instructions below. The tool does not execute agents or approve its own results.

Read the documented phase transitions and approval gates →

Verified GitHub Actions execution · 18 October 2026

What ProjectPilot returned in a real run

ProjectPilot was installed from its repository on an isolated Ubuntu runner, using Python 3.12. The commands below ran successfully on a synthetic “Review Assistant” project. This is a verified output excerpt, not a simulated terminal.

$ pp init "a local code review assistant" --name "Review Assistant"
Initialized ProjectPilot project 'Review Assistant' (review-assistant).
Current phase: idea

$ pp status
Current phase: idea
Next phase: validation
Active gate: Run `pp validate` to proceed.

$ pp next
1. Enter validation
   Priority: High
   Reason: The idea has not been validated yet.
   Suggested command: pp validate

Read full recorded CLI output →Verify successful CI execution ↗

The run creates local state and provides advice; it neither executes an AI model nor automatically approves a project.

Try the tool

Quick start

Commands adapted from the project's documented README. Run from the repository root after cloning it; consult the full README for prerequisites and platform-specific setup.

git clone https://github.com/migmcc/project-pilot.git
cd project-pilot
python -m pip install -e .
pp init "a local code review assistant" --name "Review Assistant"
pp status
pp next
pp dashboard

Installation details and full usage ↗

Scope and trust

Important limitations

It does not run models, execute skills, edit code or publish results. The latest stable release identified by its README is v1.0.0.

View on GitHub ↗Browse other projects →