Repository brief

gsd-build/get-shit-done

Read the upstream summary on the left, browse the cached forks below it, and load each fork comparison into the right-hand panel.

Cached analysis

cached 2026-03-30T11:58:46.064Z

gsd-build/get-shit-done

gsd-build/get-shit-done is a popular MIT-licensed Node.js/TypeScript CLI package for meta-prompting, context engineering, and spec-driven development across Claude Code, OpenCode, Gemini CLI, Codex, Copilot, Cursor, and Antigravity. It is actively maintained, with 45,014 stars, 3,632 forks, and recent commits through 2026-03-29. The repo is organized as an installer plus commands, agents, hooks, scripts, SDK, docs, and tests, and the package exposes `get-shit-done-cc` via `bin/install.js`.

GitHub

Stars45,014

Forks3,632

Default branchmain

Last pushed2026-03-29T12:10:03Z

Best maintainedhowlil/ez-agents

Closest to upstreamdarrylwisner/get-shit-done-github-copilot

Most feature-richnForma-AI/nForma

Most opinionatednForma-AI/nForma

Forks

Choose a fork to inspect

10 cached fork briefs

Fork comparison

howlil/ez-agents

active

significant_divergence

Choose this fork if you want a more hardened, installer-centric, and operationally simplified GSD variant and can accept being materially behind upstream. Choose upstream if you need the newest headless SDK and runtime work or want the broadest feature parity.

Likely purpose

To ship a more stable, installer-focused variant of GSD with extra runtime support, security fixes, and fewer moving parts for adopters who want the workflow without chasing upstream churn.

Best for

Teams that want GSD as a hardened, opinionated installer/runtime layer rather than a fast-moving upstream track.; Users who care more about Windows/packaging reliability and security fixes than bleeding-edge prompt/runtime features.; Adopters maintaining their own forked distribution or internal tooling around GSD.; People who prefer a simplified, professionally packaged variant of the project.

Additional features

Security hardening, including fixes for parsing-related ReDoS and regex vulnerabilities.
Installer/runtime compatibility fixes for Windows global installs and ES module/runtime folder layout.
Support work for Qwen Code installation and dist/hooks packaging.
A more professionalized README and streamlined CI/workflow setup.
Operational packaging changes such as scoped package/update behavior and prepublish/build-hook fixes.
Expands API, schema, or SDK surface area for downstream integrations and generated clients.

Missing features

It is 716 commits behind upstream, so it likely misses recent upstream additions and fixes rather than tracking them closely.
Likely lags upstream headless SDK and auto/init workflow work, including the prompt overhaul and `--sdk` installer path.
Likely lags newer runtime coverage and installer behavior added upstream in late March 2026.
Upstream docs and workflows around the latest spec-driven headless flows may not be present or fully aligned.
Some upstream behavior appears to be removed, disabled, or intentionally stripped down in this fork.
It trails upstream by 716 commits, so some recent upstream features and fixes are likely not present yet.

Strengths

More conservative and production-oriented than upstream.
Better fit for teams that value security fixes and installer reliability over rapid feature adoption.
Evidence of active maintenance and repeated quick fixes around real installation/runtime failures.
Reduced workflow sprawl compared with the upstream project's broader surface area.

Risks

Large divergence makes upstream merge-forward harder.
Being far behind upstream increases the chance of missing new CLI/runtime features and bugfixes.
Custom installer/runtime changes may behave differently from upstream documentation and support guidance.
Many changes are broad enough that adopters should expect behavioral drift and integration surprises.