Domain Apps

It structures AI agents into a three-role math modeling team—modeler, coder, and writer—enforcing stage-gate quality checks before generating competition-ready Word or LaTeX papers.

Cain is a real-world AI pentesting agent that enforces scope through engineering constraints rather than model self-discipline, with built-in coverage for AWS, Azure, GCP, and the Chinese clouds most tools ignore.

It wires Claude into a thousand trading venues so you can bet on elections, short perps, and launch Solana tokens from a single chat thread.

It exists to give retail A-share traders a lightweight, self-hosted alternative to bloated commercial terminals—no forced cloud data, no pretend AI stock picks.

It’s an open-source attempt to replace expensive SIEMs with a modular, AI-assisted security operations stack you can host yourself.

Real-time screen translation for Android games and manga, no root needed.

It spares tech transfer offices the weeks of manual patent, market, and literature review needed to assess a paper's commercial potential.

AutoHedge chains LLM-powered agents into a trading pipeline that currently works only on Solana.

Structured skill packs that teach AI agents to query AlphaGenome, UniProt, and 30+ scientific databases without burning tokens on bad requests.

freephdlabor is an open-source framework that delegates the entire research lifecycle—from hypothesis to LaTeX—to a team of coordinated LLM agents.

A Flask dashboard and Android client that repurpose e-waste into a privacy-first computer-vision pipeline with structured event logging.

To give AI agents and quant scripts a legitimate, unified pipe into official Tonghuashun A-share market data.

It exists to stop LLMs from confidently hallucinating decompiled C++ by forcing every guess through build tests and binary parity checks.

An open-source alternative to Kommo and Intercom for chat-first businesses, built to self-host AI sales agents that qualify leads and close deals natively over WhatsApp.

Xime is a deliberately minimal, Rime-based Android input method that serves as its author's personal testbed for on-device AI experiments in predictive text and speech recognition.

It stops LLMs from hallucinating fund-manager hot takes by locking them to real public statements, portfolio data, and cited method frameworks.

AlphaGBM Skills wires real-time options data and quantitative scoring into Claude Code, Cursor, and other agents so they stop making up Greeks and start reading actual volatility surfaces.

OpenWhispr is the open-source, privacy-first alternative to WisprFlow and Granola that lets you choose between local Whisper/Parakeet models or your own cloud API keys.

It gives AI coding agents the domain knowledge to read KiCad schematics, audit PCB layouts for EMC and signal integrity issues, and catch costly hardware bugs before fabrication.

It exists because most AI bidding tools charge enterprise rates that small-company staff can't afford, while the free alternatives are reportedly unusable.