jonggrang

Orchestrate AI software development through a 16-phase pipeline with persistent state.

11|2|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/porcupine-md/jonggrang --skill jonggrang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: jonggrang
Source: https://github.com/porcupine-md/jonggrang/tree/main
Command: npx skills add https://github.com/porcupine-md/jonggrang --skill jonggrang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Jonggrang provides a deterministic orchestration platform for AI-powered software development, ensuring discipline, traceability, and repeatable results in feature delivery.

Core Features & Use Cases

  • Two-mode operation (work loop and orchestrate) with a 16-phase pipeline and explicit phase mapping.
  • Two-tier skill system (core BIOS) plus library skills loaded on demand via a gateway.
  • Five-role assembly line (Lead/Developer/Reviewer/TestLead/Tester) enforced by deterministic hooks for quality gates.
  • Persistent state via MANIFEST.yaml with session resume across restarts and parallel work modes.
  • CLI-first workflow and extensible hooks, plugins, and gateways for diverse toolchains.

Quick Start

Plan a feature, decompose it into tasks, implement with AI agents, and verify results using Jonggrang.

Frequently Asked Questions about jonggrang

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run deterministic AI workflows for software development?

Deterministic AI workflows enforce strict quality gates and phase pipelines for software development. You use a CLI to plan features, decompose them into tasks, and implement them with a five-role assembly line that requires typechecks and tests before completion.

What is orchestration in AI-driven feature delivery?

Orchestration in AI-driven feature delivery coordinates a 16-phase pipeline with a five-role assembly line. It ensures discipline and traceability by using deterministic hooks to manage quality gates, moving tasks sequentially through Lead, Developer, Reviewer, TestLead, and Tester roles.

Can I resume an AI workflow orchestration after an interruption?

Yes, you can resume AI workflow orchestration after interruptions using persistent state. The system saves progress in a MANIFEST.yaml file, allowing you to restart sessions and continue parallel work modes exactly where they left off.

How do AI workflow hooks enforce code quality gates?

AI workflow hooks enforce code quality gates by binding strict validation rules to a 16-phase pipeline. They require typechecks, tests, and reviews to pass before a phase can complete, ensuring repeatable results and preventing unverified code from advancing.

Do I need a specific framework to use deterministic AI orchestration?

No specific framework is required because deterministic AI orchestration is CLI-first and supports extensible hooks, plugins, and gateways. This design allows it to integrate with diverse toolchains and load library skills on demand via a gateway.

What is the best way to manage AI agents in a software pipeline?

The best way to manage AI agents in a software pipeline is using a two-mode workflow with deterministic hooks. This approach maps explicit phases to a five-role assembly line, ensuring agents produce traceable, repeatable results under strict quality constraints.