Shipped through it so far: 4 live products and an npm package → the showcase.

Welcome to Builder Mode

Every builder knows the moment: it's 11pm, you have an idea, and your hands are itching to build. Then the "right way" kicks in — business plans, market sizing, twenty validation interviews — and by the time the strategy funnel is done with you, the spark is dead. Strategy-first kills more builders than bad ideas ever did.

That advice was written for a world where building was expensive. That world is gone. With AI, building is cheap, fast, and reversible — the thing you ship is the test.

So I've made it my mission to simplify the builder spirit into a momentum-first, strategy-second mode: Builder Mode. The idea is to enable millions to start leveraging AI to build better, build faster, and build more — to build aggressively with instinct, validate iteratively, and scale with strategy.

But most solo projects don't die in the build — they die in everything after it: the security review, the launch, pricing, first customers, knowing when to kill. Hamzaish is Builder Mode as a working system — the Claude Code setup that doesn't stop at code. It runs the whole lifecycle — Ideate → MVP → Launch → Sell → Scale → Kill-or-double-down — with stage-specific agents, playbooks distilled from real ships, and a brain that carries every lesson into your next product.

→ Read the full mission: Builder Mode and The full philosophy →

Quickstart — first win in 5 minutes

One command (installs Bun if missing, clones, sets up — read it first):

curl -fsSL https://raw.githubusercontent.com/hamza-ali-shahjahan/hamzaish/main/install.sh | sh

You need Bun and Claude Code.

git clone https://github.com/hamza-ali-shahjahan/hamzaish.git
cd hamzaish
bun run setup        # idempotent — creates YOUR factory, never touches existing data

Then open Claude Code and type:

/builder-mode a tip calculator for freelancers

Watch it scaffold a local-first product that runs in 60 seconds — zero accounts, zero config. Add your stack (Stripe, Resend, Supabase…) only when you want it with /go-live, and ship when you're ready. (The 10-minute guided version →)

Safety: scaffolded products run agent-generated code inside a devcontainer, secrets are gitignored from commit zero, and nothing auto-pushes off your machine. (Full threat model →)

What you get

• 32 agents organized by lifecycle stage — idea validation, architecture, scope-guarding, security review, landing copy, SEO, pricing, cold outreach, retention, kill-or-double-down. Not just builders: the after-the-build crew.
• A pre-launch security gate that blocks. 59 concrete checks (auth, authz, data exposure, secrets) with a forced BLOCK/CLEAR verdict.
• 39 playbooks + 128 distilled practices — launch sequencing, first-100-customers, pricing, production ops. Short, sourced, and honestly badged: proven by a real ship, or marked research-baked until it is.
• A brain that remembers. SQLite-indexed learnings, decisions, and anti-patterns, searchable from any session via /brain-ask. Your second product starts smarter than your first. (brain/)
• Sane infra defaults — Next.js + Supabase + Stripe + Resend + PostHog + Sentry, free-tier-first, pre-wired in every scaffold. (the stack, and the accounts you set up once →)
• Portfolio discipline — /portfolio-pulse across everything you run; quarterly kill-or-double-down so zombie projects don't eat your year.

Go deeper

Your first product in 10 minutes · Architecture · Philosophy · Where it's heading · Security model · Contributing · Changelog

What's inside

Everything below maps 1:1 to the repo's folders — the counts are the filesystem, not the marketing.

📚 The practices ledger (128)

BEST-PRACTICES.md — 128 practices for shipping products with Claude Code: 26 ✅ proven by real ships and dated incidents · 3 🟡 partially proven · 99 ⏳ research-baked from named sources. Anti-patterns lead — each one cost us something real. Every line links to its deep-dive playbook and its source.

🤖 The agents (32)

One router + 31 stage agents under factory/agents/. Each is a markdown SKILL.md your Claude Code session invokes by intent — the routing table lives in CLAUDE.md.

💡 Idea stage (7)

🏗️ MVP stage (5)

🚀 Launch stage (9)

📈 Scale stage (6)

🗂️ Portfolio (4) + the router

🛠️ The skills & commands (29)

17 skills + 12 commands under factory/skills/ and factory/commands/ — auto-discovered by Claude Code after bun run setup. Most commands wrap a skill of the same name; the deeper protocol lives in the skill's folder.

📖 The playbooks (39)

Short (300–800 words), sourced, stage-gated. The practices ledger is the scannable index over these.

How it's different

Now that you've seen what's inside — the side-by-side:

Architecture

brain/        — identity, principles, learnings, anti-patterns, decisions, ingested knowledge
factory/      — agents (idea/ mvp/ launch/ scale/ portfolio/), skills, commands, playbooks
products/     — one folder per product: metadata + learnings ONLY (code stays in private repos)
meta/         — changelog, retros, evals, the self-improvement loop
stack/        — tech defaults + the set-up-once accounts guide
templates/    — Next.js starter + doc templates

Product code is never in this repo — only metadata and learnings. Your code (the moat) stays private; locations are wired via a git-ignored code-paths.local.json. So the repo is safe to share without exposing anyone's secret sauce. (the public/private boundary →)

The discipline

1. Build is the default — validate before irreversible bets. Cheap, fast, reversible ships are validation. Before expensive moves: ~5 target-profile conversations. The hard rule: never skip it silently — bun run check-validation <slug> records the debt. (The wp-to-astro lesson, encoded.)
2. Scope is the moat. Every product's scope.md says what it does AND deliberately doesn't.
3. Persistent context. Every product gets a CLAUDE.md; every decision is logged in decisions/.
4. Measurement before launch. North-star, activation, retention, false-positive shape — defined before the first user.
5. The factory is a product. If it can't ship product #1 through, fix the factory before adding slots.
6. Honest copy. Every outward-facing word is true and verifiable when it ships; aspiration is labelled, never present-tense. What's proven vs. promising is tracked in the honest ledger.

The self-improvement loop

Every working session appends learnings to brain/learnings/. At major-cycle boundaries, /learn-loop scores candidates on five axes (rubric) and promotes only the top few into load-bearing guardrails — a skill rule, a playbook step, an anti-pattern, a line in the practices ledger. Quarterly, /kill-or-keep runs on Hamzaish itself and re-checks each promoted guardrail: deliver the predicted gain, or get sunset. The factory compounds; it doesn't ossify.

Budget posture: free tiers and pay-per-query by default; every subscription needs written ROI in a product's decisions/. (the bootstrapped stack →)

Built on, and informed by

The backbone is hard-won venture experience — the Business-SWAT roles, opportunities, and mentors that came with years at Disrupt.com, taking things from zero to one before AI made building cheap. On that foundation, the patterns studied and credited — Addy Osmani's spec→ship discipline, Karpathy's eval-driven flywheel, gbrain (knowledge graph), Anthropic's Founder's Playbook (lifecycle framing), hermes-agent (self-improving skills), openclaw (multi-channel gateway), and ponytail (multi-agent portability) — sharpened that instinct and 10×'d the AI and agentic-building learning on top of it. Study material lives in references/, never imported.

→ The full credits, and the roll

License

TL;DR — free for builders. Don't take it closed-source and sell it. Commercial license on request.

AGPL-3.0 — see LICENSE. Copyright © 2026 Hamza Ali.

In plain English: use, study, modify, and self-host freely. If you run a modified version as a network service, your source must be AGPL too — the factory stays open for solo builders; nobody quietly turns it into a closed product. Commercial license available for closed-source use — contact below.

Built in public by Hamza Ali — [email protected]. The factory's repo runs on the factory's own discipline.