b780cd2dc6
Epic #1671 Phase 6 "Done when" required a maintainer-reachable proof that a single-fragment edit propagates to every emitted per-runtime artifact with no second source surface needing an edit. No test referenced that surface at all. Adds tests/fragment-single-edit-propagation.install.test.cjs (20 rows): a hard-linked overlay repo overrides exactly ONE steps/ fragment, real installers are spawned per runtime, and the emitted artifacts are asserted directly. Expected runtime sets derive from RUNTIME_META at run time, never a hardcoded count, so a new runtime cannot be silently under-covered. Six negative controls keep it from being pass-always theater. An identity-stubbed composer must make marker bytes LEAK, proving the marker-absence assertion can fail. Each derived generator whose --check is used as evidence has its own red-path control driven by an override-only edit, each asserting the generator's own typed reason enum rather than matching prose. Coverage is disclosed, not implied. REGEN_STEPS_WITHOUT_CHECK_MODE names the regen:derived steps with no read-only mode; CONTENT_EDIT_INSENSITIVE_CHECKS names gen-inventory-manifest, whose --check derives from directory listings and is structurally blind to content edits. Both constants are pinned by a test so the disclosure cannot silently rot. Assertions check sentinel PRESENCE, not whole-file byte identity: partial-wave.md embeds the runtime-launcher snippet, so emitted fragments are legitimately rewritten per runtime (windsurf -> .windsurf, qwen -> .qwen, claude -> its absolute config dir). A dedicated row now locks that behavior in. The overlay tree-diff is labelled a harness self-check, not the no-cascade proof it cannot be: the overlay is built from the checkout with the override map applied, so that diff can only ever restate the test's own fixture. Extracts buildOverlayRepo into tests/helpers/overlay-repo.cjs so both install suites share one implementation instead of diverging copies, converts that sibling's six try/finally test bodies to t.after() per CONTRIBUTING.md, and frees each per-runtime temp install eagerly so peak disk stays bounded. Refs #2933 Co-authored-by: sim <sim@local>
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.