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/TOPSHARK JIM/GSTACK

topshark-jim/gstack

1 skill · 1 installs · GitHub

Install

npx skills add https://github.com/topshark-jim/gstack

Skills in this repo

1Plan Eng Reviewgstack plan-eng-review is a procedural agent skill that forces a thorough execution-plan review before any implementation starts—aimed at solo builders who otherwise jump straight from a vague spec to codegen. The workflow behaves like an engineering manager review session: it stress-tests architecture and data flow, insists on diagrams where helpful, surfaces edge cases and performance risks, and maps test coverage expectations. A explicit priority hierarchy tells the agent never to skip Step 0 or the test diagram even when context is tight. Recommendations are meant to be opinionated—calling out DRY violations, under-engineering, and over-abstraction—while still pausing for your decision instead of silently choosing a path. It respects repo-local instructions in AGENTS.md, TODO.md, and TODOS.md when present. Use it when you have a plan document or task breakdown and want a human-in-the-loop gate that reduces rework, flaky tests, and missing failure modes. The outcome is a locked, reviewed plan you can hand to implementation skills with clearer scope and acceptance signals.1installs