ralph

Orchestrate autonomous feature development from prd.json with checks before commits.

Updated Jul 8, 2022
One-click install
npx skills add https://github.com/limaon/dotfiles --skill ralph-limaon
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ralph
Source: https://github.com/limaon/dotfiles/tree/main/.config/opencode/skills/ralph
Command: npx skills add https://github.com/limaon/dotfiles --skill ralph-limaon

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jq, droid, and includes scripts (resource) components.

What problem does it solve?

This Skill automates end-to-end feature development by orchestrating an autonomous agent that reads a PRD (prd.json) and executes user stories, enabling rapid, repeatable delivery without manual orchestration.

Core Features & Use Cases

  • Autonomous loop: reads product requirements, creates/updates prd.json, and executes user stories one at a time.
  • Memory persistence: stores progress learnings in progress.txt and maintains state across iterations via prd.json; archives prior runs on branch changes.
  • Safety and traceability: commits are made only after successful checks; progress is appended to progress log for auditing.

Quick Start

Run the agent in the current project: ~/skills/ralph/scripts/ralph.sh [max_iterations]

Frequently Asked Questions about ralph

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

FAQPage Schema
How do I automate end-to-end feature development from a product requirements document?

Autonomous end-to-end feature development works by orchestrating an AI agent that reads prd.json, executes user stories sequentially across isolated iterations, and updates the PRD upon completion. It handles the full delivery loop without manual orchestration.

How does an autonomous agent maintain progress and memory across multiple iterations?

The agent persists memory and state across iterations by appending learnings to progress.txt and maintaining execution status within prd.json. Prior runs are automatically archived when the active branch changes.

Do I need jq and droid installed to run autonomous feature implementation?

Yes, you need both jq and droid installed as dependencies. The autonomous feature implementation scripts require these tools to parse JSON requirements and orchestrate the agent workflow.

What is the best way to ensure autonomous code commits pass typecheck, lint, and test checks?

The best way to ensure autonomous commits pass checks is to use a controlled workflow that enforces typecheck, lint, and test validation before allowing any commit. The agent only commits after successful checks.

What happens to previous progress logs when switching to a new branch for feature development?

When switching to a new branch, prior runs are automatically archived to preserve history. The agent maintains traceability by appending progress to a log file and updating the PRD state.