gsd-core
GSD Core is a set of instructions that adds a phased planning and verification workflow to AI coding agents for developers building software projects.
What it is good for
- Keeping an AI coding agent focused across a large software project
- Planning and verifying each phase of a software build before shipping
- Sharing project memory between AI coding sessions
- Running research and execution in fresh subagent contexts
At a glance
- Who it is for
- Setup
- Adds to an AI app you already use
Level 2 of 5. Connects to Claude, ChatGPT, or a similar app you already have, usually through a settings screen.
- Cost
- Free and open source
- Type
- Skill or plugin
- Tasks
- Works with
- Claude Code, OpenCode, Antigravity CLI, Kimi CLI, Kilo, Codex
Momentum
- Rank this week
- Not ranked (last #43, week 43, 2026)
- Best rank
- #27 week 40, 2026
- GitHub stars
- 11,144
GitHub numbers as of Nov 2, 2026. Updated weekly.
Ranking history
Official links
Setup steps are on the official pages. We do not repeat them here, since they change.
Questions
- Is gsd-core free?
- Yes. It is free and open source (MIT license).
- Do I need to know how to code to use gsd-core?
- No. Connects to Claude, ChatGPT, or a similar app you already have, usually through a settings screen.
- What does gsd-core work with?
- According to its own documentation, it works with Claude Code, OpenCode, Antigravity CLI, Kimi CLI, Kilo, and Codex.
- What is gsd-core good for?
- Keeping an AI coding agent focused across a large software project, planning and verifying each phase of a software build before shipping, sharing project memory between AI coding sessions, and running research and execution in fresh subagent contexts.
Similar agents
- planning-with-files
planning-with-files is a skill that keeps a written plan on disk so AI coding agents can resume long tasks after context loss for developers and power users.
- compound-engineering-plugin
Compound Engineering is a plugin of 36 skills that makes each coding task leave notes for the next one, for developers using AI coding agents.
- aidlc-workflows
aidlc-workflows is a set of workflow rules that makes AI coding agents follow structured, approval-gated steps for building software.
- unlazy
unlazy is a skill that makes AI agents finish substantial work by writing checkable acceptance gates and verifying results before reporting done, for people who give agents larger multi-step jobs.
- tstack
tstack is a set of agent skills that take a pull request from change to merge for developers using Claude Code or Codex.
- Vibe Coding Toolkit
Vibe Coding Toolkit is a set of instructions and prompts that sets up AI coding agents and project foundations for developers building with AI.