supastarter vs Makerkit vs supastack: which is best for agencies?
It depends on surfaces and seats. supastarter ($349+, up to ~$1,499) is the mature web-only pick across Next.js, Nuxt and TanStack Start. Makerkit ($299+) suits teams all-in on Supabase. supastack ($199–$799) is the only one shipping mobile (Expo), desktop (Tauri) and an MV3 extension scaffold alongside two web engines — with 693+ tests, .ai/ agent guardrails, and unlimited client commercial projects on every tier including the $199 Starter.
Three Serious Kits, Three Different Bets
All three are paid, production-grade SaaS starters with real teams behind them — this is a good problem to have. They diverge on what 'full-stack' means: supastarter bets on web framework choice, Makerkit bets on the Supabase stack, and supastack bets that your product will outgrow the browser.
We maintain separate deep-dives on each pair — supastack vs supastarter, supastack vs Makerkit, and Makerkit vs supastarter. This piece is the agency-eyed summary.
Pricing: $199–$799 vs $299–$1,499
Agency economics live and die on the license terms, not the sticker price. Here is how the three line up on entry cost, top tier, and what you may build for clients:
| Boilerplate | Entry tier | Top / agency tier | Client project rights |
|---|---|---|---|
| supastack | $199 Starter (1 seat) | $799 Agency (10 seats) | Unlimited commercial client projects on every tier |
| supastarter | $349 Solo | Team tiers up to ~$1,499 | Per-license terms — check tier limits |
| Makerkit | $299+ | Higher tiers for teams | Client use depends on tier |
Platforms: Web-Only vs Four Surfaces
This is where the comparison stops being close. If a client asks for an iOS app in month three, two of these kits start a second codebase and one keeps shipping:
| Capability | supastarter | Makerkit | supastack |
|---|---|---|---|
| Web framework | Next.js, Nuxt or TanStack Start (pick one edition) | Next.js | Next.js 16 + TanStack Start — both included |
| iOS / Android | No | No | Expo mobile app included |
| Desktop | No | No | Tauri 2 — macOS, Windows, Linux |
| Browser extension | No | No | MV3 companion scaffold |
| ORM / database | Prisma or Drizzle | Drizzle (Supabase-first) | Drizzle + Postgres |
| Payments | 5 providers | Stripe-first | 8 gateways + RevenueCat mobile IAP |
| Automated tests | Unit & integration | Playwright E2E suite | 693+ tests, credential-free local gate |
| AI agent guardrails | Basic agent docs | Limited | AGENTS.md, .ai/ docs, ~29 skills, @supastack/mcp |
| Developer platform | Typed API (oRPC) | API + webhooks | Workspace API keys, signed webhooks, audit log, OpenAPI |
Agency Economics: The Rows That Actually Matter
Three licensing realities decide profitability for client work. First, projects-per-license: supastack permits unlimited commercial client end products on every tier — including the $199 Starter — so a single purchase amortizes across your entire book of business.
Second, seats: Full Bundle ships 3 seats and Agency ships 10, so the developers touching client repos are covered without per-seat negotiation. Third, delivery surfaces: when the same license already includes mobile and desktop, 'can you also do the app?' becomes an upsell instead of a subcontract.
Built for AI Agencies & Agentic Workflows (BMAD, Claude Code, Cursor)
A growing share of supastack buyers run AI software agencies on it. BMAD-style planning agents (Analyst, PM, Architect, QA) turn a client brief into scoped stories; Claude Code and Cursor implement them against a repository that already answers auth, billing, tenancy and testing; AGENTS.md, .ai/ docs and Agent Skills keep every agent on the existing patterns.
Whether you hand a client a web MVP in five minutes from a single-platform repo, or deliver iOS, Android and desktop builds from the unified monorepo, one license lets you build unlimited commercial client projects — no per-project fee, one seat per developer who needs repository access.
- Single-platform repos for fast web-only client deliverables; the unified monorepo for iOS + Android + Desktop engagements
- AGENTS.md, .ai/ file docs and Agent Skills that keep BMAD, Claude Code and Cursor agents inside the architecture
- A credential-free local test gate that catches agent regressions before your client sees them
- One license, unlimited client projects — buy on a client's behalf and keep the margin on every engagement
Verdict: Pick by Surface Map, Not Feature Grid
All three will serve a web SaaS well. The honest recommendation matrix:
- See the tiers and seat counts on the pricing page before you commit — the license math matters more than the sticker.
| Your situation | Pick |
|---|---|
| Web-only product, want the most established kit or a Nuxt codebase | supastarter |
| All-in on Supabase, want Postgres RLS done for you | Makerkit |
| Agency shipping client SaaS, any chance of mobile/desktop/extension | supastack |
| Solo founder watching budget who still wants client rights | supastack Starter — $199 |
FAQ: supastarter vs Makerkit vs supastack
What agency buyers ask when comparing the three kits:
- Is supastack really cheaper than supastarter?, At entry, yes — Starter is $199 vs $349 Solo, and the $299 Full Bundle undercuts supastarter's top tiers while adding mobile, desktop and extension.
- Which is best if my client insists on Supabase?, Makerkit — it is built Supabase-first with deep Postgres RLS. supastack runs on Postgres + Drizzle and can deploy on Supabase, but Makerkit's defaults lean further into the ecosystem.
- Can I deliver client projects on the cheapest supastack tier?, Yes — unlimited commercial client end products are permitted on every tier, including Starter; seats are the only variable (1 / 3 / 10).
- Do any of the three ship a browser extension?, Only supastack includes an MV3 companion scaffold wired to the same REST API and workspace auth.