Skip to content
/ pilot Public

Make your AI assistant remember, learn, and follow a structured approach to every task. Built for Kiro CLI.

License

Notifications You must be signed in to change notification settings

requix/pilot

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

22 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

PILOT - Self-Learning Engineering Assistant for Kiro

Structured methodology. Persistent memory. Continuous learning.

PILOT extends Kiro with:

  • The Algorithm — OBSERVE → THINK → PLAN → BUILD → EXECUTE → VERIFY → LEARN
  • Memory — Captures solutions and applies them to future work
  • Identity — Knows your goals, projects, and working style

The Algorithm

Every task follows seven phases:

  1. OBSERVE — Understand before acting
  2. THINK — Generate approaches
  3. PLAN — Select strategy
  4. BUILD — Define success criteria
  5. EXECUTE — Do the work
  6. VERIFY — Test against criteria
  7. LEARN — Capture verified insights

Core principle: Define what success looks like BEFORE you execute.

Requirements

  • Kiro installed
  • macOS or Linux (bash required)
  • (Optional) Enable semantic search: kiro-cli settings chat.enableKnowledge true

Quick Start

git clone https://github.com/requix/pilot.git
cd pilot
./install.sh

Run: kiro-cli --agent pilot

What to Expect

After install, PILOT will:

  • Follow the Algorithm for complex tasks
  • Remember solutions that work
  • Learn your preferences over time

Credits

Inspired by PAI by Daniel Miessler.

License

MIT License — Use freely, commercially or personally.

About

Make your AI assistant remember, learn, and follow a structured approach to every task. Built for Kiro CLI.

Topics

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages