# Benchtop Labs

> The automated quality verification standard and curated registry for modern open source developer tools, libraries, and utilities.

## Overview
Benchtop Labs is an open quality standard for independent developer software. Every listed project undergoes a transparent, automated 6-point bench test suite run inside an isolated sandbox environment. Results are published openly with zero partial credits, no sponsored placements, and no bypass mechanisms.

Website: https://benchtoplabs.vercel.app
Public Registry & Intake: https://github.com/Ameya79/benchtoplabs-registry
Specification: https://github.com/Ameya79/benchtoplabs-registry/blob/main/BENCH_SPEC.md

## The 6-Point Bench Test Suite
1. README (`readme`): Clear documentation explaining architecture, usage examples, and configuration.
2. OSI License (`license`): Recognized open-source license file in repository root matching SPDX standards.
3. Semantic Release (`release`): Tagged semantic version release (`vX.Y.Z`) on GitHub Releases.
4. Active Maintenance (`activity`): Commits to default branch within the last 365 days.
5. Sandbox Install (`installs`): Clean, isolated installation in an ephemeral environment (pip/npm/cargo/go/binary).
6. Passing CI (`tests`): Continuous integration workflows on default branch are 100% green.

## Verified Projects on the Bench
All founding projects currently on the bench pass 6 of 6 checks (100% pass rate):

### Cuttle
- Description: Right-click any file in Windows Explorer, pick the format you want, and get a converted copy instantly.
- Ecosystem: binary
- Authors: Ameya Kulkarni
- Bench Score: 6 of 6 passed
- Developer Installs: Verified v1.0 Release
- Install: `curl -sLO https://github.com/Ameya79/cuttle/releases/latest/download/Cuttle-Setup-x64.exe`
- Spec Tag URL: https://benchtoplabs.vercel.app/p/cuttle
- GitHub Repository: https://github.com/Ameya79/cuttle
- Official Website: https://cuttle-website-gamma.vercel.app/
- Documentation: https://cuttle-website-gamma.vercel.app/
- Package Release: https://github.com/Ameya79/cuttle/releases/latest
- README Badge: `[![Bench Tested](https://benchtoplabs.vercel.app/api/badge/cuttle)](https://benchtoplabs.vercel.app/p/cuttle)`

### rollit
- Description: Rolling window statistics for numpy arrays.
- Ecosystem: pypi
- Authors: Ameya Kulkarni
- Bench Score: 6 of 6 passed
- Developer Installs: 1,300+ Installs
- Install: `pip install rollit`
- Spec Tag URL: https://benchtoplabs.vercel.app/p/rollit
- GitHub Repository: https://github.com/Ameya79/rollit
- Official Website: https://rollit-website.vercel.app/
- Documentation: https://rollit-website.vercel.app/
- Package Release: https://pypi.org/project/rollit/
- README Badge: `[![Bench Tested](https://benchtoplabs.vercel.app/api/badge/rollit)](https://benchtoplabs.vercel.app/p/rollit)`

### Talanton
- Description: Automatic spending limits on OpenAI, Claude, and 60+ AI models.
- Ecosystem: pypi
- Authors: Ameya Kulkarni
- Bench Score: 6 of 6 passed
- Developer Installs: 1,300+ Active Installs
- Install: `pip install "talanton-py[openai]"`
- Spec Tag URL: https://benchtoplabs.vercel.app/p/talanton
- GitHub Repository: https://github.com/Ameya79/Talanton
- Official Website: https://talanton-py.vercel.app/
- Documentation: https://talanton-py.vercel.app/
- Package Release: https://pypi.org/project/talanton-py/
- README Badge: `[![Bench Tested](https://benchtoplabs.vercel.app/api/badge/talanton)](https://benchtoplabs.vercel.app/p/talanton)`

### Mustel
- Description: Non-AI static analysis layer for AI coding agents with token-efficient diagnostics.
- Ecosystem: pypi
- Authors: Ameya Kulkarni & Raunak
- Bench Score: 6 of 6 passed
- Developer Installs: 6,000+ Installs
- Install: `pip install mustel`
- Spec Tag URL: https://benchtoplabs.vercel.app/p/mustel
- GitHub Repository: https://github.com/mustel-py/mustel
- Official Website: https://mustel.vercel.app/
- Documentation: https://mustel.vercel.app/
- Package Release: https://pypi.org/project/mustel/
- README Badge: `[![Bench Tested](https://benchtoplabs.vercel.app/api/badge/mustel)](https://benchtoplabs.vercel.app/p/mustel)`

### Driftwood
- Description: Stateless Monte Carlo stock price simulation and quantitative risk engine API in Python.
- Ecosystem: other
- Authors: Ameya Kulkarni
- Bench Score: 6 of 6 passed
- Developer Installs: 1,000+ Risk Simulations
- Install: `git clone https://github.com/Ameya79/Driftwood.git`
- Spec Tag URL: https://benchtoplabs.vercel.app/p/driftwood
- GitHub Repository: https://github.com/Ameya79/Driftwood
- Official Website: https://driftwood-docs.vercel.app/
- Documentation: https://driftwood-docs.vercel.app/
- Package Release: https://github.com/Ameya79/Driftwood
- README Badge: `[![Bench Tested](https://benchtoplabs.vercel.app/api/badge/driftwood)](https://benchtoplabs.vercel.app/p/driftwood)`

## Additional Resources for AI Agents & Developers
- Full Detailed Manifest (all check outputs, dependencies, package sizes): https://benchtoplabs.vercel.app/llms-full.txt
- JSON API: https://benchtoplabs.vercel.app/api/projects
- Submit a Project: https://github.com/Ameya79/benchtoplabs-registry/issues/new?template=submit-project.yml
- Sitemap: https://benchtoplabs.vercel.app/sitemap.xml