63404ca45b
Thirty-eight tests, covering every previously unreached branch in this module: malformed and non-object config at each nesting level, the origin/HEAD path returning a bare remote prefix, the remote-show parse missing its HEAD line and its "(unknown)" sentinel, three catch arms, the spawn-failure path that reports unverified, the worktree probe answering something other than "true", a non-zero toplevel query, a blank toplevel, and the default diagnostic sink on both arms. The one worth naming is the local-branch fallback that was deleted earlier today as unreachable and restored. It is hard to pin because it and the empty-stdout guard above it both return null, so the return value cannot tell them apart. The test drives it through a result object whose stdout is a getter that counts reads: one read means the guard fired, two means the guard passed and the fallback ran. Deleting the line again fails the value assertion — the function would return undefined — and changing the guard to swallow whitespace fails the read count, which is the other way that branch dies. Tier attribution is now pinned. Four tests rig the lower tiers to answer with DIFFERENT branch names and assert the recorded call sequence, so a result from symref, remote-show, local-branch, or config can no longer be mistaken for another. Previously any of them could have produced the answer and the test would not have noticed. Each tier's argv and timeout are asserted exactly. The boundary trio applies to the branch listing rather than a numeric limit: zero matching lines, one, and two. Three assertions already in this file were removed rather than left beside the new ones. A doesNotThrow around the worktree probe became an exact-value test of the catch arm; a typeof-string-and-non-empty shape check on the resolved root became an equality check against the realpath; and the unverified-fallback diagnostic was asserted only to be non-empty, so it now asserts the exact sentence, which fails if the reason changes rather than only if it vanishes. Two folded bodies used try/finally for teardown and now use t.after. Refs #3057 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
GSD Core
Git. Ship. Done.
English · Português · 简体中文 · 日本語 · 한국어
A light-weight meta-prompting, context engineering, and spec-driven development system for Claude Code, OpenCode, Antigravity CLI, Kimi CLI, Kilo, Codex, Copilot, Cursor, Windsurf, and more.
What is GSD Core
GSD Core is a context-engineering and spec-driven development framework that drives AI coding agents (Claude Code, Codex, Antigravity CLI, Kimi CLI, Copilot, Cursor, and more) through a disciplined phase loop. It solves context rot — the quality degradation that accumulates as an AI fills its context window — by running all heavy research, planning, and execution work in fresh-context subagents while keeping your main session lean.
How it works
Each milestone repeats the same five-step loop, one phase at a time:
- Discuss — capture implementation decisions before anything is planned
- Plan — research, decompose, and verify the plan fits a fresh context window
- Execute — run plans in parallel waves; each executor starts with a clean 200k-token context
- Verify — walk through what was built; diagnose and fix before declaring done
- Ship — create the PR, archive the phase, repeat for the next one
Quickstart
npx @opengsd/gsd-core@latest
The installer prompts for your runtime (Claude Code, OpenCode, Antigravity CLI, Kimi CLI, Kilo, Codex, Copilot, Cursor, Windsurf, and more) and whether to install globally or locally. The installer is required for cross-runtime compatibility — do not copy files from agents/ or commands/ directly.
On another runtime or without Node.js? See Install on your runtime.
Once installed, start a new project or onboard an existing repo:
/gsd-new-project # greenfield project
/gsd-onboard # existing codebase
New here? Follow Your first project for a guided walkthrough from install to first shipped phase, or Onboarding an existing codebase for brownfield setup.
Documentation
What's new in 1.7.0 → docs/whats-new-1.7.0.md
Tutorials — learning by doing:
How-to guides — task-focused recipes:
Reference — authoritative facts:
Explanation — concepts and design decisions:
Full index: docs/README.md. Other languages: 日本語 · 한국어 · Português · 简体中文.
Why it works
Most AI-coding setups fail at scale because context bloat silently degrades output quality, there is no shared memory between sessions, and nothing verifies that code actually works. GSD Core solves all three: heavy work runs in fresh subagents, structured artifacts like STATE.md and CONTEXT.md survive session boundaries, and the verify step walks through what was built and generates fix plans before a phase is declared done. See docs/explanation/context-engineering.md for the full reasoning.
Troubleshooting? See docs/how-to/recover-and-troubleshoot.md.
Community
| Project | Platform |
|---|---|
| gsd-opencode | Original OpenCode port |
| Discord | Community support |
Star History
License
MIT License. See LICENSE for details.
Claude Code is powerful. GSD Core makes it reliable.