ff161f2281
* fix(#1582): derive phase-complete velocity from By-Phase table (idempotent) updatePerformanceMetricsSection blind-added summaryCount onto the prior velocity total on every phase complete, so re-running phase complete on an already-complete phase incremented the total each time (the sibling of #4, which fixed the Completed Phases counter the same way). The velocity total is now derived as the sum of the By-Phase table's Plans column AFTER the row upsert — re-completing a phase upserts the same row, so the sum is stable; a hand-edited inflated total self-heals downward to the true sum on the next completion. When the By-Phase table is absent the total is left unchanged (no crash). Strengthens the misnamed 'idempotent' test (its comment explicitly declined to assert velocity idempotency — the latent gap) and adds a self-heal regression; corrects the #320 behavior-lock velocity assertion which had encoded the blind-add (3 = 1+2 double-count) — the derived value is 2. * chore(#1582): backfill changeset pr ref to 1655 * fix(#1582): velocity sum tolerates indented By-Phase rows (codex review) Adversarial review (codex, gpt-5.5/high) flagged that byPhaseTablePattern's data-row capture allows leading whitespace ([ \t]*\|), but the derive sum was anchored at ^\| and would skip indented hand-edited/legacy rows — capturing them in the table but silently undercounting. Align the sum regex (^\s*\|) with the table capture's tolerance. Adds an indented-row regression. Two other codex findings are pre-existing and out of scope: padded/unpadded phase dedup (phaseRowPattern, identical in old code — derive yields the same value as the old blind-add) and CRLF tables (the shared byPhaseTablePattern header requires bare \n, so the upsert was already broken on CRLF; the fix changes stale-vs- double-count, does not worsen it). * fix(#1582): add verification fixtures to velocity tests under #1522 gate Post-rebase onto next+#1548, the #1582 velocity tests (self-heal, indented-row) use phase complete, which now fail-closes under #1522's canonical verification gate without a passed *-VERIFICATION.md. Add writePassedVerification(tmpDir,'02-next','02') to both.
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, Gemini 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, Gemini 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, Gemini 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 your first project:
/gsd-new-project
New here? Follow Your first project for a guided walkthrough from install to first shipped phase.
Documentation
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.