Skip to content

Skillselionwith God's help

By phase

IdeaResearch & spot demand
ValidateScope & prototype
BuildFrontend, backend, MCP
ShipTest, review, deploy
LaunchDistribution, SEO, ASO
GrowAnalytics & retention
OperateMonitoring & uptime

By type

Skills
Marketplaces
MCP servers
Workflows

Skills Marketplaces MCP Learn Brief

Browse all tools →By phase

Idea Validate Build Ship Launch Grow Operate

Explore

Skills
Marketplaces
MCP
Learn
Brief

This week for builders

Five minutes, every Monday — the tools, releases and tactics for shipping solo.

Email address

unsubscribe anytime.

Skillselion

An independent, curated directory of AI-coding tools for solo builders - skills, MCP servers, marketplaces and workflows for Claude Code, Codex and Cursor, organized by the indie build journey from idea to operate.

ResourcesBrowse Skills Browse MCP Servers Browse Marketplaces Browse Workflows Best-of shortlists Glossary State of AI Skills 2026 Weekly Brief

CommunityAbout Learn Feedback Privacy Policy Terms Cookie Policy Accessibility Cookie settings

Last updated June 10, 2026 · © 2026 Skillselion. Skillselion is an independent project and is not affiliated with Anthropic, OpenAI, Cursor, Claude, Claude Code, or Codex.

Skills/PAGEAI PRO/RALPH LOOP

pageai-pro/ralph-loop

1 skill · 1.5k installs · 253 stars · GitHub

Install

npx skills add https://github.com/pageai-pro/ralph-loop

Skills in this repo

1Prd Creatorprd-creator is an agent skill that converts a completed product requirements document into simple, concise, actionable, and verifiable implementation tasks stored as JSON. It expects the PRD at PROJECT_ROOT/.agent/prd/PRD.md, preserves unique PRD task identifiers as TASK-${ID} entries, and always inserts TASK-1 for prerequisite verification before feature work. Output lands in a task index at PROJECT_ROOT/.agent/tasks.json and detailed specs under PROJECT_ROOT/.agent/tasks/TASK-${ID}.json, which solo builders and agent loops can execute step by step. The workflow explicitly pauses after PRD creation to ask whether to generate tasks, then interviews the user when any task remains ambiguous. It fits indie builders running Ralph-style agent loops who want PRD fidelity carried into machine-checkable work units rather than a vague todo list in chat.1.5kinstalls