Voyage Digital
    Tools· AI Assistance & Agent Building

    GSD (Get Shit Done)

    GSD is an open source workflow and orchestration framework for AI coding agents, using a five phase cycle of Discuss, Plan, Execute, Verify and Ship to prevent context loss over long sessions.

    planningstructuringexecutableOpen Source
    Source code
    npx @opengsd/gsd-core@latest, asks on its own which runtime environment to target. No manual copying from agents/ or commands/.
    Open provider
    GSD (Get Shit Done)

    Description

    GSD addresses the quality loss that occurs when an AI model fills its context window over a long session. The framework structures work into five phases instead: Discuss (capture decisions), Plan (research and break work into steps), Execute (implement plans in parallel waves), Verify (review and fix issues) and Ship (create the pull request, archive the phase). Sub agents each start with a fresh context window rather than inheriting the buildup of a long main session. According to the source, GSD is optimized in particular for Claude Code, but also works with OpenCode, GitHub Copilot, Cursor and other harnesses.

    Each project gets its own planning structure with project, roadmap and status files, against which individual phases are planned and checked off. Adopting GSD across a project means adopting its folder and file conventions as well, and stepping away from it later means a corresponding cleanup effort. As an actively developed open source project, its conventions can change between versions.

    Strengths

    Five phase structure
    Each milestone follows the same five step cycle according to the source, Discuss, Plan, Execute, Verify and Ship, one phase at a time.
    Fresh context per execution wave
    Each executor starts with a clean 200,000 token context window according to the source, rather than the buildup of a long main session.
    Parallel execution waves
    Plans are implemented in parallel waves according to the source instead of one step after another.
    Cross harness
    Usable according to the source with, among others, Claude Code, OpenCode, Antigravity CLI, Kimi CLI, Kilo, Codex, Copilot, Cursor and Windsurf.
    Guided project start
    Separate commands for a new project and for onboarding an existing codebase, plus tutorials, how-to guides, reference and explanation documentation.

    Assessment

    AI features

    • Discuss phase Captures implementation decisions according to the source before anything is planned.
    • Plan phase Researches and decomposes the task and checks according to the source whether the resulting plan fits a fresh context window.
    • Execute phase Implements plans in parallel waves according to the source, each executing agent with its own clean 200,000 token context.
    • Verify phase Walks through what was built according to the source, diagnoses issues and fixes them before a phase is considered done.
    • Ship phase Creates the pull request according to the source, archives the phase and hands off to the next one.

    Suitable for

    • Teams and individuals working with AI coding agents on multi phase projects who want to avoid context loss over long sessions
    • Projects that need a reproducible, phase based workflow from discussion through planning to verification and pull request
    • Users of multiple coding harnesses who want a workflow standard that works across tools
    • Less suitable for Less suited to very small, one off changes, where the phase overhead is out of proportion to the task

    Limitations and notes

    • Phase overhead on small tasks For very small, one off changes, the effort of the five phases is out of proportion to the task.
    • Adopting project wide conventions Anyone who introduces GSD also adopts its folder and file conventions for project, roadmap and status files; stepping away later means a corresponding cleanup effort.
    • Changing conventions As an actively developed open source project, its conventions can change between versions.
    • Installer instead of manual copying According to the source, the project explicitly advises against copying files directly from agents/ or commands/; only the installer ensures compatibility across multiple runtime environments.

    Quick start

    1. Run the installer: `npx @opengsd/gsd-core@latest`. According to the source, it asks on its own which runtime environment to target (for example Claude Code, OpenCode, Codex, Cursor) and then whether to install globally or locally.
    2. Run `/gsd-new-project` for a new project.
    3. Run `/gsd-onboard` for an existing codebase.
    4. To get started, follow the "Your first project" tutorial, which according to the source walks through everything from install to the first shipped phase; for brownfield setups there is "Onboarding an existing codebase".

    Tips

    • For problems, check docs/how-to/recover-and-troubleshoot.md first, the repository links its own guide for this.
    • Before changing agent or command files, rerun the installer instead of copying manually, otherwise compatibility across runtime environments is lost according to the source.
    • For non-English-speaking teams, the README according to the source is also available in Portuguese, Simplified Chinese, Japanese and Korean.

    Access

    Repository: https://github.com/open-gsd/gsd-core (license: MIT). According to the README, installation runs through the npx installer npx @opengsd/gsd-core@latest, which asks on its own which runtime environment to target (among others Claude Code, OpenCode, Codex, Cursor, Windsurf). The project explicitly advises against manually copying files from agents/ or commands/. We provide access to the source here only, with no hosting of our own.

    Last reviewed: · Pricing, plans and features are a snapshot in time. Check the provider's own page before deciding.

    In the workshop this becomes your method.

    Whoever sees this process run once wants the agent behind it next. We build that in the workshop From Process to Agent.

    View workshops

    Related resources

    Browse all resources
    Skills

    The Role Rulebook

    Guides a project through five clearly separated roles, Explorer, Planner, Builder, Reviewer and Navigator, so the actual problem is understood before anything gets planned, built or reviewed.

    Learn more
    Assistants

    The Assistant Curator

    Turns your briefing into a complete, ready-to-use AI assistant configuration, including system prompt, knowledge sources, tool selection and a reasoned model recommendation.

    Learn more
    EV
    Tools

    everything-claude-code (ECC)

    everything-claude-code (ECC) is an MIT-licensed meta-collection of skills, agents, slash commands, hooks and rules packs that standardizes agentic coding workflows across multiple AI coding harnesses.

    Learn more
    skillsmith
    Tools

    skillsmith

    skillsmith is a meta-suite for building, auditing and distilling consistent Claude Code skills across four steps: Discover, Scaffold, Distill and Audit.

    Learn more
    Skills

    The Skill Builder

    Guides the creation and iterative improvement of a skill, from clarifying intent through test runs to fine-tuning the description that triggers it.

    Learn more
    Assistants

    The Context Interviewer

    A guided interview in five blocks that turns your answers into a complete context profile, covering company, role, team, priorities and communication style, ready to use as a knowledge base for any further AI assistant.

    Learn more
    Assistants

    The Persona Curator

    Builds a concrete persona from interviews, usage data, and observations that an AI agent can adopt and embody directly, complete with a ready-to-use short description for its own system prompt.

    Learn more
    Assistants

    The Prompt Interviewer

    A guided interview that turns your answers on goal, audience, output and rules into a ready-to-use system prompt with a reasoned LLM recommendation.

    Learn more
    Agentforce
    Tools

    Agentforce

    Salesforce's platform for autonomous AI agents in sales, support, marketing and commerce, deeply integrated into the Salesforce Data Cloud.

    Learn more
    ChatGPT
    Tools

    ChatGPT

    OpenAI's AI assistant built around GPT-5, with Canvas for collaborative editing, an Agent Mode for browser tasks, and image and video generation through DALL-E 3 and Sora.

    Learn more
    Claude
    Tools

    Claude

    Anthropic's AI assistant family (Opus, Sonnet, Haiku), used as a core tool throughout the Voyage Digital stack for development and operations, strong at coding, long documents, and agentic workflows via MCP servers.

    Learn more
    Claude Managed Agents
    Tools

    Claude Managed Agents

    Anthropic's managed cloud infrastructure for running Claude as an autonomous, tool-using agent without building your own runtime.

    Learn more
    Copilot 365
    Tools

    Copilot 365

    Microsoft's AI assistant, integrated directly into Word, Excel, PowerPoint, Outlook and Teams, designed for companies already working within the Microsoft 365 ecosystem.

    Learn more
    Grok
    Tools

    Grok

    Grok is xAI's AI assistant with real-time access to X, a multi-step DeepSearch research function, image generation, voice mode, and task-based research agents, particularly strong for current events and trend research.

    Learn more
    Langdock
    Tools

    Langdock

    Enterprise AI platform from Berlin that gives teams central, governed access to multiple language models, Custom Assistants and no-code agents under GDPR compliant EU hosting.

    Learn more
    LibreChat
    Tools

    LibreChat

    LibreChat is a self-hostable open-source chat frontend that unifies multiple AI providers, agents and MCP servers in a single interface with single sign-on and user management.

    Learn more
    Lindy
    Tools

    Lindy

    Lindy is a no-code platform for building AI agents that handle email, meetings, lead qualification and customer support for you.

    Learn more
    Open WebUI
    Tools

    Open WebUI

    Open WebUI is a self-hostable, offline-capable interface for local and remote language models, with Ollama integration, local RAG and enterprise features such as SSO and audit logs.

    Learn more
    Relevance AI
    Tools

    Relevance AI

    Relevance AI is a no-code platform for building autonomous AI agents for sales, support and operations, with an enterprise focus and an EU hosting option.

    Learn more

    Conversation, not pitch

    Understand first, then decide. We take time for an initial conversation, without sales pressure, without obligation.

    Schedule a call