Log in Get Started
CLI Tool

create-stacksfinder

Deterministic stack recommendations.
No AI hallucinations. No account required.

npm bun pnpm
npx create-stacksfinder@latest my-app
84 Technologies Scored Across 8 categories
6 Criteria
<5ms Scoring Time
$0 Forever Free
01

What create-stacksfinder does

A single command that recommends and scaffolds an optimized tech stack based on your project requirements.

Deterministic Scoring

Same inputs always produce the same outputs. No AI randomness. Every recommendation is reproducible, auditable, and explainable.

Core principle

Offline-Capable

All data embedded. Works without internet.

Project Generation

Scaffolds JavaScript/TypeScript projects: Next.js, SvelteKit, Nuxt, Remix, Astro, and more.

Instant Results

Scoring completes in under 5ms.

Technology Coverage

84 technologies • 8 categories
Frameworks 11
Next.jsSvelteKitNuxtRemixAstro+6
Databases 12
PostgreSQLSupabaseNeonMongoDB+8
ORMs 8
DrizzlePrismaKysely+5
Auth 4
Better AuthAuth.jsClerkAuth0
Hosting 9
VercelRailwayFly.ioCloudflare+5
Payments 4
StripePaddleLemonSqueezyPolar
Backend 10
HonoElysiatRPCFastify+6
Rust 8
AxumActixRocketDiesel+4
02

How the scoring engine works

Understanding the algorithm helps you trust the recommendations. Here's exactly what happens when you run the CLI.

▸ SCORING_PIPELINE.exe
DETERMINISTIC <5ms
INPUT
01

Load Scores

Versioned data source

02

Apply Weights

High/low priorities

03

Apply Deltas

Constraint modifiers

04

Compatibility

Tech synergies

05

Assign Grade

A+ → C- rank

A +
OUTPUT

The 6 Scoring Criteria

Weight distribution Σ = 100%

Developer Experience

Docs, TypeScript, tooling

20%

Ecosystem

Community, packages, jobs

20%

Maintainability

Upgrades, LTS, stability

20%

Performance

Speed, cold starts, bundles

15%

Cost

Hosting, licensing, ops

15%

Compliance

Security, audits, residency

10%
Scoring Principles
CORE
1 //Weighted composite of all criteria
2 //Bounded adjustments prevent gaming
3 //Same inputs → same outputs, always
∑ Composite Weighted
⊂ Adjustments Bounded
= Output Deterministic
03

What it does NOT do

Transparency matters. Here's what the CLI deliberately avoids.

Privacy & Trust Guarantees

No Cloud Required

Zero API calls. Your project context never leaves your machine.

100% Local

Minimal Telemetry

Anonymous stats only. Opt-out: DO_NOT_TRACK=1

Opt-out Available

No AI Hallucinations

LLMs never select technologies. Pure algorithmic scoring only.

Deterministic

No Rate Limits

Generate 100 projects. Script it. No server, no limits.

Unlimited

No Dark Patterns

No popups, no "hit your limit" messages. Just a CTA at the end.

Ethical UX

Open Source Scoring Data

The scoring data is visible in the npm package. Audit exactly how recommendations are made.

View on npm
04

Why it's 100% free

$0 Forever

Free tier should be genuinely useful, not a crippled demo.

Our pricing philosophy
?
Why give it away?
01

Genuinely Useful

Free tools should solve real problems, not be crippled demos. The CLI delivers complete recommendations you can act on.

Philosophy
02

Zero Marginal Cost

It runs on your machine. We don't pay for compute. No economic pressure to restrict.

Sustainable
03

Community Feedback

More users = more edge cases = better data. The community improves the product.

Win-Win
No hidden limits. Use it 1,000 times. Script it. CI/CD it. We don't throttle free users.

Ready to try it?

One command. Zero setup. Instant recommendations.

npx create-stacksfinder@latest my-app
View on npm