Open Source
I love contributing to open source — it's how I learn, give back, and collaborate with developers around the world. Everything below is pulled live from GitHub.
289
Merged PRs
108
Open PRs
121
Repositories
406.4k
Stars reached
Notable projects
A few widely-used projects I've landed changes in.
All contributions
A progressive Node.js framework for building efficient, scalable, and enterprise-grade server-side applications with TypeScript/JavaScript 🚀
Open-source AI job search: scan job portals, evaluate listings into a structured A-H report with a global 1-5 score, tailor your CV, track applications — runs locally in your AI coding CLI (Claude Code, Codex, OpenCode, Antigravity…)
Orca is the ADE for working with a fleet of parallel agents. Run any coding agent with your own subscription. Available on desktop, mobile and VPS.
The uncompromising Python code formatter
Event Driven Orchestration & Scheduling Platform for Mission Critical Applications
Python Development Workflow for Humans.
EmDash is a full-stack TypeScript CMS based on Astro; the spiritual successor to WordPress
Statsmodels: statistical modeling and econometrics in Python
Build your own AI SRE agents. The open source toolkit for the AI era.
🏹 Better dates & times for Python
Reads key-value pairs from a .env file and can set them as environment variables. It helps in developing applications following the 12-factor principles.
Retrying library for Python
The Sphinx documentation generator
✨ The fundamental numerical library for JavaScript and TypeScript. ✨ ⭐️ Star to support our work!
N-D labeled arrays and datasets in Python
Aesthetic, minimalist platform for learning Japanese inspired by Duolingo and Monkeytype, built with Next.js and sponsored by Vercel. Beginner-friendly with plenty of good first issues - all contributions are welcome!
Neo.mjs is a self-evolving software organism: a professional end-to-end AI engineering team whose cross-model swarm inhabits live apps via Neural Link, Active Hybrid GraphRAG, DreamService, and self-healing loops.
Django-environ allows you to utilize 12factor inspired environment variables to configure your Django application.
📚 The open-source, offline-first Notion, Obsidian & Confluence alternative. Advanced Markdown, multi-tenant teams, OIDC/SSO, and local S3 backups. Deploy in one command.
Coverage plugin for pytest.
pytest plugin for distributed testing and loop-on-failures testing modes.
Run-time type checker for Python
Tools for publishing transcripts for Claude Code sessions
Settings management using pydantic
An extended commonmark compliant parser, with bridges to docutils/sphinx
Style-preserving TOML library for Python
Serve static websites (SPA's) using Nest framework (node.js) 🥦
Chaos and resiliency testing tool for Kubernetes with a focus on improving performance under failure conditions. A CNCF sandbox project.
Local evidence debugger and trajectory-test toolkit for TypeScript AI agents: inspect causal runs, catch wrong tool paths in CI, and share safe offline evidence.
Monocle is a framework for tracing GenAI app code. This repo contains implementation of Monocle for GenAI apps written in Python.
Self-hosted runtime control plane for AI agents. Observe or HITL approve or Block rogue tool calls before it executes: secret leaks, prompt injection, supply chain etc in a local dashboard. Agent agnostic (Claude, codex, langchain etc.)
|toqito> (Theory of Quantum Information Toolkit) is a Python library for research in quantum information theory.
Open-source Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO) platform. Track how ChatGPT, Claude, Perplexity, Gemini, Copilot, Grok, and Google AI Overviews mention and cite your brand in AI search. MIT-licensed, self-hostable, every metric auditable in code.
A sphinx extension for designing beautiful, screen-size responsive web components.
Pytest plugin for testing examples in docstrings and markdown files.
Apache Maven Archetype (Plugin)
Project Management for the agent era
CI-native security testing for MCP servers. Attack simulation, schema drift detection, and health scoring before agents depend on them.
CAD CLI and MCP server for AI agents
Codex skill for evidence-preserving academic rewriting in Chinese and English.
High performance, multimodal-native engine for AI workloads.
mloda.ai - Open Data Access for AI and ML. Plugin-based. Traceable. Framework-agnostic.
MCP server for predictive maintenance and machinery fault diagnosis. Gives AI assistants evidence-based vibration analysis - FFT, envelope, bearing fault detection, ISO 20816-3 severity - with a measured, blind CWRU benchmark. Local-first: raw signals never leave your machine. Includes a Claude Code plugin.
A Python framework for modular, self-contained skill management for machines.
DoHabit is a lightweight, local-first habit tracker designed to respect your privacy. No registration, no ads, no cloud tracking - just open it and start building your routine instantly.
A terminal AI agent you can fully undo - every file and shell action is snapshotted and reversible. Model-agnostic (any LLM), and connects to 1000+ app tools and MCP servers.
Formats your tox.ini files
Check for differences between requirements.txt files and your environment
Scheduled, unattended Claude Code CLI jobs — runs on your Claude subscription via the local CLI, no ANTHROPIC_API_KEY.
A local-first, AI-powered image search engine. Find your photos using natural language queries—100% private and self-hosted.
Private, offline-first PDF and image tools that run 100% in your browser. No uploads, no sign-up. Now on the Chrome Web Store.
Fast, zero-dependency access log analyzer, security threat inspector, and TUI dashboard for Caddy v2
CONTINUUM: Verifiable semantic recovery for long-running AI agents. Semantic checkpoints (not conversation dumps), an idempotent action ledger that refuses duplicate side effects, and a hash-chained tamper-evident event log, all exposed as a deny-by-default MCP server. Framework-agnostic, Python 3.11+.
inplan — a Markdown editor for human ⇄ coding-agent planning. CLI + desktop editor.
Open-source Python toolkit for water data, hydrology, and agricultural water management — 29 unified collectors (USGS, UK EA, Hub'Eau, GRDC, BOM, FAO…), Bulletin 17C flood frequency, FAO-56 ET₀, an MCP server, and an AI methodology recommender.
Self-hosted mission control for real Claude Code and Codex sessions — persistent terminals, automations, and browser/mobile control.
Get help exactly when and where you need it. Grafana Interactive Learning brings contextual documentation and interactive tutorials directly into Grafana, so you can learn and build without leaving your workflow.
Single-pass exploratory data analysis on streaming algorithms. Bounded memory at any dataset size, pandas and Polars, self-contained HTML reports.
Code-First AI Compliance. The open-source regulatory engine for the EU AI Act. Automate your risk assessments directly within your CI/CD pipeline.
Making programmatic programmable
Free, open source video workflow
Open-source no-progress governor for AI coding agents. Detect repeated work that changes nothing. Shadow first, enforcement only after evidence.
Stack traces that tell the tale — a Logback appender that turns Java errors into AI-ready reports
A cozy, safety-first desktop Git client, with a cat mascot and a Safety Manager that snapshots before every mutation so Undo is always one keystroke away
Open Knowledge Graphs and Ontologies
《Random Dice 2 - Wiki》 fan-made
Split incoming payments between multiple recipients on Stellar. Soroban contracts, TypeScript SDK, and dashboard.
High five or [hive hive] is a project that aims to gain insights into wild bees.
A local, AI-powered CLI that lets you chat with free LLMs (via OpenRouter) to edit files, run shell commands, and manage git—all from your terminal.
A leakage-verified prompt-injection benchmark for measuring and improving agent robustness (defensive).
Discord bot that watches BookMyShow and DMs you the moment tickets open. Self-hosted with one command. Bun + TypeScript + SQLite, no API keys.
Instagram comment-to-DM automation you actually own. One Cloudflare Worker, one D1 database, $0/month. Self-hosted ManyChat alternative.
Performance engineering and observability for AI agents.
Registry of official plugins for mloda.ai
Metaorcha : many models, one harness. Apache 2.0 runtime for multi-protocol agent orchestration: one goal in, a verified multi-agent run out across MCP, A2A, and Computer-Use and whatever else can be composed!
Does concept-injection introspection emerge with scale? A faithful, controlled reproduction charted across model-size ladders.
Open research on local-first, model-agnostic long-term memory for LLM agents
Leakage-verified contextual-integrity benchmark: does an agent disclose a sensitive attribute to a recipient the context forbids. Defensive, synthetic-only. Sibling of leakgauge.
The steering-vector report card: effect size, side effects, dose-response, and per-layer sensitivity for concept vectors.
A contract-first Python framework for turning typed endpoint declarations into executable APIs.
Find, explain, and safely fix bad data. Local-first data quality for files, databases, CI pipelines, and AI agents.
Modern SOAP library for Python. Build servers with a SoapService class and soap_operation decorators (auto-generated WSDL), or drive a typed client from an existing WSDL. SOAP 1.1/1.2, all binding styles, WS-Security, MTOM, WS-Addressing. ASGI/WSGI-compatible. Hardened lxml parser and tests conformance suite.
Real-time YouTube watch party application built with React, TypeScript, Node.js, Socket.IO, and MongoDB. Watch videos together with synchronized playback and live chat.
Research platform for discovering and rigorously falsifying crypto trading strategies: event-sourced paper trading, implementation-parity verification, and a documented negative-results record.
Local-first CLI: which Claude Code skills, subagents and hooks actually fire, what they cost, and which are dead weight.
Typed Python scene generator for multi-camera perception. Fluent DSL for cameras, motion, occlusion, and human pose; emits calibrated N-camera manifests with ground truth. The producer that feeds multicam-occlusion.
Open-source Vietnamese web accessibility scanner and WCAG checker for developers.
Domain-specialized privacy / contextual-integrity leakage benchmark for health and fintech. Synthetic-by-design, verified disclosure, no LLM judge in scoring.
Template repository for plugin creation for mloda.ai
Evidence-backed engineering health analysis for public GitHub repositories. Every score is explainable, every finding is traceable to public GitHub data.
पाऊस · Hyperlocal Mumbai 0–2h rain nowcast. Calibrated free forecasts, public scoreboard vs METAR VABB. $0 to run.
Reproducible matplotlib wrapper with mm-precision layouts
Run surveys from your terminal. Agent-friendly, type-safe, OSS.
Contract event indexer for Stellar/Soroban. Stores events durably past the RPC's retention window, and serves them over a JSON API. The core of the SoroTrail stack: SoroScope is the explorer UI on top of it, SoroBeacon the monitoring and alerting layer.
Dependency-free Python static analyzer for code health, complexity, security risks, and import cycles
Turn a web crawl into an ML-ready dataset. Boilerplate extraction, Gopher/C4 quality filtering, MinHash near-dedup, sharded output. Zero dependencies.
Turn a folder of markdown notes into a queryable knowledge graph — people, companies, projects and the relationships between them — so Claude answers from your actual notes instead of guessing. A Claude skill + two Python scripts. No API key, no token, cross-platform.
Docs for Tuna OS
Natural-hazard alerts from 59 official alerting authorities, normalised into one CAP-shaped schema with provenance. Relays official warnings; never issues them.
Find duplicate files
Universal scientific data I/O with a plugin format registry
Open-source SafeRide mobile app, owned API/local infrastructure, website, and public-safe on-device AI tooling.
An end to end LLM work flow platform
CPU ANN retrieval engine with a public, reproducible performance competition framework
Always-on-top desktop panel for Windows 11 showing your Claude subscription usage live - 5-hour window and weekly limit, burn rate, and end-of-week projection. Works offline from the local log, or signs in to claude.ai for all devices. 12 languages.
Static AI-agent hook security audit with bounded config-to-script provenance—without executing scanned commands.
Map dependencies, cycles, hotspots, and potential change blast radius in one local-first CLI.
Compose complementary local AI-agent capabilities with explainable lexical coverage and explicit gaps.
Local-first AI-agent trace analysis with a high-confidence Recovery Ledger for serial retry evidence.
MCP Server to integrate with Millennium Dawn's mod development tool kit.
Manage Matter devices in a Smarthome
Open-source harmonic tide visualization for 24 coastal locations worldwide.
Template repository for an npm package (TypeScript + tsup + Jest + ESLint + Changesets + GitHub Actions).
A static analyzer and linter for VRChat UdonSharp scripts
Catch OpenAPI breaking changes and security-sensitive contract drift before they ship.