Language
HomeLanguages
Sections
A review of tick-stock-panel, an open-source quant analysis platform with Polars and DuckDB that runs locally in Docker. Includes backtesting, factor mining, and LLM integration.
A lightweight Rust-based web scraper that uses just 14MB of RAM and outputs clean Markdown for LLMs.
Browser Harness connects LLMs directly to your real Chrome browser via CDP, letting agents write missing helper functions on the fly for web automation tasks.
An open-source self-hosted terminal for stock screening, backtesting, and monitoring. Built with Polars, DuckDB, and FastAPI.
Apache Maka brings event sourcing to AI agents, recording every action to an append-only SQLite log for full auditability and reproducibility.
AI-Infra-Guard is an open-source platform for securing AI services and agents. It scans AI infrastructure for CVEs, audits MCP servers and agent skills, and tests LLM resistance to jailbreaks.
OpenEcon's awesome-ai-for-economists repository brings together AI tools, MCP servers, and econometric libraries tailored for rigorous data work.
Freehire parses job listings directly from ATS systems, bypassing aggregators. With 3.3M vacancies and a full open-source stack, it's a complete candidate workspace.
Zero is a Go-based terminal AI coding agent that runs locally with your choice of models—no vendor lock-in, strict sandboxing, and full session control.
Modern social media recommendation algorithms stopped working for users. OpenBiliClaw is a local AI agent that gathers interests from multiple platforms, builds a psychological profile, and searches for content independently.
A practical guide to fine-tuning 8B models on consumer hardware using Soup—a CLI tool that brings layer streaming and automated pipelines to small GPUs.
Delegating Google Dorking to neural network agents: a look at DorkAgent, a tool that automates search queries using Serper API and LLMs.
Discover how Ix transforms your codebase into a queryable dependency graph, saving 30-99% of tokens by replacing blind grep with targeted graph queries.
Cali is a CLI tool that uses LLM agents to automate mobile testing, performance profiling, and code review for React Native and Expo projects.
Skales is a local AI agent for desktop and mobile that runs autonomously on your system, reads files, manages your browser, and executes task chains without constant oversight.
Anyone building AI agents in production knows the pain of duct-taping five different tools together. Future AGI aims to solve this with one open-source platform.
Open-source alternative to OpenAI Deep Research that runs locally with Docker, supports custom LLM providers, and costs nothing beyond your API usage.
Standard Markdown renderers weren't built for live token streams. Markstream solves flickering, broken code blocks, and hanging formulas when rendering AI output.
Hyper-Extract converts messy documents into structured knowledge graphs and typed data using ready-made templates, eliminating the need for complex prompt engineering.
OpenKnowledge opens local Markdown folders with a full WYSIWYG editor and built-in AI agent integration via MCP. No proprietary databases, just plain files and Git sync.
Unity-MCP connects Unity with Claude and Cursor via the Model Context Protocol, giving AI assistants direct access to scenes, assets, and the Roslyn compiler.
Engineer Andrey Mikhailov built TurboFieldfare, a custom Swift/Metal runtime that runs Google's Gemma 4 26B-A4B model in just 2 GB of RAM on a basic MacBook Air M2.
Omnigent is an open-source Python meta-framework that orchestrates multiple AI coding agents under one roof with unified interfaces and security policies.
holaOS combines multiple AI agents in one desktop environment with shared context, local memory, and interactive panels for seamless workflow.