playbooks

home / skills / ratacat / claude-skills / lfg

lfg skill

/skills/lfg

This skill automates an end-to-end engineering workflow, orchestrating planning, review, testing, and video generation to accelerate delivery.

npx playbooks add skill ratacat/claude-skills --skill lfg

Review the files below or copy the command above to add this skill to your agents.

Files (1)
SKILL.md
577 B
---
name: lfg
description: Full autonomous engineering workflow
---

## Arguments
[feature description]

Run these slash commands in order. Do not do anything else.

1. `/ralph-wiggum:ralph-loop "finish all slash commands" --completion-promise "DONE"`
2. `/workflows:plan $ARGUMENTS`
3. `/compound-engineering:deepen-plan`
4. `/workflows:work`
5. `/workflows:review`
6. `/compound-engineering:resolve_todo_parallel`
7. `/compound-engineering:test-browser`
8. `/compound-engineering:feature-video`
9. Output `<promise>DONE</promise>` when video is in PR

Start with step 1 now.

Overview

This skill orchestrates a full autonomous engineering workflow for building, testing, and delivering features from source materials. It combines planning, iterative development, parallel task resolution, browser-based testing, and multimedia feature delivery to produce PR-ready artifacts. The goal is repeatable end-to-end feature delivery with minimal human coordination.

How this skill works

The skill sequences planning, deepening, work execution, review, parallel TODO resolution, browser testing, and feature video generation. It inspects source materials and an existing plan, refines tasks, executes engineering steps, runs tests in a headless browser environment, and produces a short demonstration video attached to the pull request. The workflow emits a simple completion promise when the video and PR artifacts are ready.

When to use it

  • Automating end-to-end feature development from specification to PR
  • Scaling feature creation across multiple small tasks or microservices
  • Validating UI changes with automated browser tests and recordings
  • Delivering clear, demo-ready pull requests with supporting media
  • Reducing manual coordination for routine engineering workflows

Best practices

  • Provide concise, well-structured source materials and acceptance criteria up front
  • Break large features into focused sub-tasks to enable parallel resolution
  • Include reproducible test steps and deterministic data for browser tests
  • Keep video demos short (30–90 seconds) focused on acceptance points
  • Review generated outputs promptly and iterate on the plan for future runs

Example use cases

  • Turn a product spec into a runnable plan, implement feature branches, and open a PR with tests and demo video
  • Parallelize TODO items across agents to speed up feature completion
  • Run automated browser smoke tests after code changes and attach a recorded failure demo to the PR
  • Create a reproducible demo video to accompany a release or feature showcase
  • Use for rapid prototyping where end-to-end artifacts (code, tests, demo) are required quickly

FAQ

What inputs does the workflow need?

Supply a clear feature description, acceptance criteria, and any relevant source materials or repo references so the planner can create actionable tasks.

Can tasks run in parallel?

Yes. The workflow supports parallel resolution of independent TODO items to accelerate completion while preserving review checkpoints.

How is the demo video generated?

The workflow runs browser-based tests and records a short feature video demonstrating acceptance criteria; that video is attached to the PR as part of delivery.

What signals completion?

Completion is signaled when the PR contains code, tests, review artifacts, and the feature demo video, at which point the workflow emits a final completion token.