Gan Build — Development skill for Claude Code
Run a generator/evaluator build loop for implementation tasks with bounded iterations and scoring.
How to install Gan Build
Installs to ~/.claude/skills/mturac-everything-openai-codex-gan-build/SKILL.md
mkdir -p ~/.claude/skills/mturac-everything-openai-codex-gan-build && curl -fsSL https://raw.githubusercontent.com/mturac/everything-openai-codex/HEAD/commands/gan-build.md -o ~/.claude/skills/mturac-everything-openai-codex-gan-build/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Gan Build does
description: Run a generator/evaluator build loop for implementation tasks with bounded iterations and scoring.
Parse the following from $ARGUMENTS:
brief— the user's one-line description of what to build--max-iterations N— (optional, default 15) maximum generator-evaluator cycles--pass-threshold N— (optional, default 7.0) weighted score to pass--skip-planner— (optional) skip planner, assume spec.md already exists--eval-mode MODE— (optional, default "pl
Alternatives in Development
- Ralph Loop — Autonomous coding loop — works through tasks, commits, repeats 14k ★
- Domain Driven Design — Apply DDD patterns and bounded contexts 663 ★
- Gyoshu Auto — Start goal-based autonomous research with bounded execution 244 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Adversarial Dev
GAN-inspired three-agent harness that pits a generator against an adversarial evaluator to build applications
Loop Execution Evaluator
Verify implementation quality. Evaluate-Loop Step 4.
Gap Loop
Loop post-implementation gap analysis until 2 consecutive clean rounds (bounded at 6, per-finding 2-fix limit,
Reconcile Tasks
Bounded, schedulable reconciler that fixes issues sitting in the wrong state against a fixed, enumerated rule
Storeops Harness Dotnet
StoreOps API (.NET 8 / ASP.NET Core) governed by a working Claude Code harness — Planner, Generator, Evaluator
Unity Ralph
Relentless verify-fix loop — refuses to stop until the project is clean. Runs unity-verifier repeatedly with c
Related Agents
Gan Evaluator
GAN Harness — Evaluator agent. Tests the live running application via Playwright, scores against rubric, and p
Gan Planner
Expand a one-or-two-sentence brief into acceptance criteria, a file list, and a 5-row scoring rubric. First st
Synthetic Org Generator
Generates the synthetic Xray (tests, plans, executions, runs) and Azure DevOps (epics/features/stories/tasks,