02a36d3db9
full test (windows-latest, 24, shard 1/3) has been RED on next since ac1b6d679 and blocks
every PR. Three assertions fail, Windows-only, and all three are stale TESTS rather than
resolver defects.
Two of them create a BARE extensionless 'fallow' and assert it resolves. #3618 deliberately
stopped resolving that on Windows, and said so in its own commit message: the extensionless
file is npm's POSIX sh shim, which CreateProcess cannot run (#3275). The code did what it
meant to; the tests were never updated to match.
Both now assert BOTH platform contracts rather than skipping a lane — the precedent #3618
set when it fixed its own F4 ('a t.skip on one lane would have been green and would have
left the win32 carve-out unpinned'). POSIX keeps the original bare+chmod fixture verbatim.
win32 creates the artifact npm actually writes there, fallow.cmd, and asserts resolution
finds it; each also pins that a bare extensionless fallow ALONE still resolves to null on
win32, so the carve-out is pinned rather than merely stepped around. The precedence test
keeps testing precedence: on win32 both node_modules/.bin and the PATH dir get a
fallow.cmd, and .bin must still win.
The third failure was case, not logic: the test wrote fallow.cmd and asserted exact string
equality, but candidates are built by appending PATHEXT entries, and PATHEXT is uppercase.
Confirmed at the source — DEFAULT_PATHEXT = '.EXE;.CMD;.BAT;.COM'
(src/shell-command-projection.cts:651), read verbatim and never lowercased, with the
candidate returned as-is. So the resolver returned fallow.CMD, which is the same file on
case-insensitive NTFS. Fixed the assertion to compare case-insensitively on win32; the
resolver is untouched, because its return value has to stay usable verbatim.
Not verifiable here: this repo's remote runner matrix is Linux-only and the host is macOS,
so the win32 branches are reasoned from the resolver source and proven only by GitHub CI.
The POSIX lanes were verified by execution (bare+chmod in .bin resolves, .bin beats PATH,
non-executable in PATH yields null).
No changeset: changeset-lint's USER_FACING_PREFIXES does not include tests/, so this is
OK_NO_USER_FACING_CHANGES rather than a missing fragment.
Refs #3618
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.