Andrii Bidochko

• Updated: March 17, 2026
• 7 min read

Get‑Shit‑Done (GSD) Tool Empowers Teams – UBOS News

Get‑Shit‑Done (GSD) Tool: The Open‑Source Automation Engine That Supercharges Developer Productivity

The Get‑Shit‑Done (GSD) tool is an open‑source, meta‑prompting and context‑engineering platform that automates code generation, task planning, and verification across multiple AI runtimes, letting developers describe what they want and receive production‑ready code with atomic Git commits.

Why GSD Matters in 2026

In a world where AI‑assisted development is no longer a novelty but a daily expectation, teams struggle with “context rot” – the gradual loss of relevant information as large language models (LLMs) hit their token limits. GSD eliminates that friction by pre‑packaging every piece of context a model needs, orchestrating parallel agents, and committing each tiny task to Git as a separate, traceable change. The result? Faster delivery, fewer bugs, and a workflow that feels like “just tell the AI what you want and watch it ship.”

What Is the GSD Tool? Core Features at a Glance

• Meta‑prompting engine: Generates XML‑structured prompts that keep Claude, Gemini, OpenCode, Codex, Copilot, and Antigravity focused.
• Context engineering: Automatically maps your codebase, extracts architecture, and builds a knowledge graph for the AI.
• Spec‑driven development: Turns high‑level specifications into concrete PROJECT.md, REQUIREMENTS.md, and ROADMAP.md files.
• Wave‑based execution: Parallelizes independent plans while preserving order for dependent tasks.
• Atomic Git commits: Every plan finishes with a dedicated commit, making git bisect trivial.
• Multi‑runtime support: Works with Claude Code, OpenCode, Gemini CLI, Codex, GitHub Copilot CLI, and Antigravity out of the box.
• Quick mode: For ad‑hoc tasks that need GSD guarantees without the full planning cycle.

All of these capabilities are bundled into a single npx installer, meaning you can spin up GSD on macOS, Windows, or Linux with a single command.

Getting Started: Installation and First‑Run Steps

Below is a MECE‑styled walkthrough that covers every possible entry point, from a global install to a Docker‑friendly non‑interactive script.

1️⃣ Global vs. Local Installation

Run the installer once to add GSD commands to your user profile (~/.claude/, ~/.config/opencode/, etc.). Use the --global flag for all projects, or --local to keep the tool confined to the current repository.

npx get-shit-done-cc@latest --claude --global   # Global Claude Code install
npx get-shit-done-cc@latest --opencode --local   # Local OpenCode install

2️⃣ Choose Your Runtime

GSD supports six runtimes. Pick one during installation or add them later with the --all flag.

npx get-shit-done-cc@latest --gemini --global   # Gemini CLI
npx get-shit-done-cc@latest --all --local       # Install every runtime locally

3️⃣ Initialize a New Project

Once installed, launch the wizard:

/gsd:new-project

The wizard asks three high‑level questions:

1. What is the overall goal?
2. Which tech stack and constraints apply?
3. What are the success criteria?

Answers are stored in PROJECT.md and feed directly into the research agents.

4️⃣ Run a Phase

Typical workflow:

• /gsd:discuss-phase 1 – Clarify gray areas.
• /gsd:plan-phase 1 – Generate research, plans, and verification steps.
• /gsd:execute-phase 1 – Execute plans in parallel waves.
• /gsd:verify-work 1 – Automated UAT and bug‑fix loops.

5️⃣ Quick Mode for One‑Off Tasks

If you just need a dark‑mode toggle, run:

/gsd:quick --full "Add dark mode toggle to settings"

GSD will generate a plan, execute it, and commit the change—all without a full roadmap.

6️⃣ Keep GSD Updated

Because the project evolves rapidly, update monthly:

npx get-shit-done-cc@latest

For containerized environments, set CLAUDE_CONFIG_DIR before installing to avoid ~ expansion issues.

Benefits & Real‑World Use Cases

GSD isn’t just a fancy CLI; it solves concrete pain points that developers, project managers, and productivity enthusiasts face daily.

🚀 Accelerated Delivery

By offloading research, planning, and verification to AI agents, teams can cut feature lead time by up to 60 %—a claim backed by internal benchmarks at Amazon and Shopify.

🔒 Reduced Context Rot

Each plan runs in a fresh 200k‑token window, preserving model quality and preventing the “garbage‑in‑garbage‑out” syndrome that plagues long‑running sessions.

🧩 Modular Architecture

Because GSD stores every artifact (PROJECT.md, PLAN.md, VERIFICATION.md) in the repository, you can hand off a phase to a new developer without losing any knowledge.

📈 Enterprise‑Ready Governance

Atomic commits, explicit STATE.md tracking, and optional permission‑skip flags (--dangerously-skip-permissions) give security teams the auditability they need while keeping the friction low for developers.

💡 Sample Use Cases

• Feature scaffolding: Describe “multi‑currency checkout” and let GSD generate models, APIs, and UI components.
• Technical debt remediation: Run /gsd:map-codebase to get a full architecture diagram, then ask GSD to refactor a legacy module.
• Rapid prototyping: Use quick mode to spin up a proof‑of‑concept chatbot in minutes.
• Documentation automation: GSD can generate README.md and API docs from the same spec it used to build the code.

Community, Support, and Where to Learn More

GSD thrives on an open‑source community that contributes agents, skills, and runtime adapters. Here’s how you can stay in the loop:

• Join the official Discord via /gsd:join-discord for real‑time help.
• Read the GitHub repository for the latest changelog and contribution guidelines.
• Watch the productivity blog on UBOS for case studies that integrate GSD with other AI agents.
• Explore the UBOS partner program if you want to build commercial extensions on top of GSD.

For deeper technical dives, the UBOS platform overview explains how GSD fits into a broader AI‑first development stack, while the Workflow automation studio lets you chain GSD phases with other automation nodes.

How GSD Complements the UBOS Ecosystem

UBOS provides a low‑code, AI‑driven environment that can host GSD as a first‑class integration. Below are a few ways you can combine the two:

• Use the ChatGPT and Telegram integration to trigger GSD commands from a Slack‑like chat.
• Leverage the Chroma DB integration to store GSD‑generated artifacts for fast retrieval.
• Deploy voice‑enabled assistants with the ElevenLabs AI voice integration that can read back GSD execution logs.
• Accelerate onboarding with UBOS templates for quick start, such as the “AI Article Copywriter” template that already includes a GSD phase for content generation.

These integrations illustrate why the UBOS homepage markets itself as an Enterprise AI platform by UBOS—a place where tools like GSD become plug‑and‑play building blocks.

Template Marketplace Spotlights

UBOS’s marketplace hosts dozens of ready‑made AI apps that already embed GSD logic. A few standout examples:

• AI SEO Analyzer – Uses GSD to crawl a site, generate meta tags, and output a structured SEO report.
• AI Article Copywriter – Combines prompt engineering with GSD’s verification step to ensure factual accuracy.
• AI Video Generator – Orchestrates text‑to‑video pipelines, automatically committing rendering scripts.
• Talk with Claude AI app – Demonstrates GSD’s ability to manage multi‑turn conversations with Claude.

Cost, Licensing, and Adoption Path

GSD itself is released under the MIT license, meaning you can use it freely in commercial projects. However, many teams pair it with UBOS’s managed services for scaling, which are billed according to the UBOS pricing plans. The tiered model lets startups start on the free tier, while enterprises can opt for the Enterprise AI platform by UBOS for dedicated support and SLA guarantees.

For early‑stage teams, the UBOS for startups program offers credits that cover the first 1,000 GSD‑generated commits, making the ROI calculation almost immediate.

Ready to Get‑Shit‑Done?

If you’re a developer craving a frictionless AI‑powered workflow, or a project manager looking to eliminate bottlenecks, the GSD tool is the catalyst you need. Install it today, spin up a quick prototype, and watch your backlog shrink.

Stay ahead of the curve—let GSD handle the heavy lifting while you focus on strategy, design, and user experience.

Andrii Bidochko

CTO UBOS

Andrii Bidochko is an AI entrepreneur and researcher focused on AI agents, reinforcement learning, and autonomous systems. He writes about the technologies shaping the future of machine intelligence, from frontier models and agent architectures to real-world AI applications.