Get Shit Done (GSD): Meta-Prompting System for Reliable AI Code Generation
A new tool called Get Shit Done (GSD) is tackling one of the biggest problems in AI-assisted development: context rot — the quality degradation that happens as coding assistants fill their context windows during long sessions.
The Problem It Solves
"Vibecoding" — describing what you want and letting AI generate code — has a reputation for producing inconsistent results that fall apart at scale. GSD acts as a context engineering layer that makes AI coding assistants reliable over longer development sessions.
How It Works
Rather than adding enterprise-style ceremony (sprint planning, story points, retrospectives), GSD uses:
- Meta-prompting — structured prompts that give the AI everything it needs to do the work and verify it
- Context engineering — manages context window utilization to prevent quality degradation
- Subagent orchestration — breaks complex tasks into manageable pieces
- State management — tracks progress across sessions
- Spec-driven development — you describe your idea, the system extracts requirements, and the AI builds it
Supported Runtimes
GSD works across multiple AI coding assistants:
- Claude Code (primary)
- OpenCode (open source, free models)
- Gemini CLI
- Codex (skills-based installation)
- GitHub Copilot CLI
- Antigravity (Google, Gemini-based)
Installation
npx get-shit-done-cc@latest
The installer prompts for runtime selection and installation scope (global or per-project). Engineers at Amazon, Google, Shopify, and Webflow are reportedly using it.
Philosophy
"I'm not a 50-person software company. I don't want to play enterprise theater. I'm just a creative person trying to build great things that work." — TÂCHES, creator
The complexity lives in the system, not in your workflow. Behind the scenes: context engineering, XML prompt formatting, subagent orchestration, and state management. What you see: a few commands that just work.
Source: GitHub - gsd-build/get-shit-done