This fork rebuilds a popular multi-agent trading framework for the A-share market, replacing Yahoo Finance with free local data feeds and adding analysts who track policy shifts, hot-money flows, and lockup expirations instead of just fundamentals.
Domain Apps
underdogs breaking outOpen Science is a local, model-agnostic desktop app that lets AI agents run code, query scientific databases, and leave behind a reproducible trail of artifacts instead of just chat transcripts.
OfficeDex exists because copying AI output into Word and reformatting it is still too much manual labor.
It gives AI agents a structured toolkit to generate CAD geometry, describe robots, and hand off files for fabrication and simulation.
It chains a dozen LLM agents into a visual workflow so your on-call engineer only has to tap 'approve' instead of SSHing in at 3 AM.
An open-source desktop alternative to Claude Science that runs the full research loop—exploration, experiments, papers, and provenance—without leaving your local folders by default.
It bridges Claude Code to your local TradingView Desktop app through the Chrome DevTools Protocol, letting an AI read charts, edit Pine Script, and navigate timeframes without touching TradingView's servers.
AiSOC is an open-source, self-hostable AI SOC that logs every LLM prompt and tool call so you can replay investigations instead of trusting a black box.
Because your coding assistant shouldn't need a securities API cheat sheet to analyze A-shares.
Pits PatchTST, TFT, and N-HiTS against CatBoost to help energy markets pick a forecasting model.
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.
Vocalinux is a fully offline, GPLv3 voice dictation app that pipes transcribed text into any Linux application on X11 or Wayland.
This PowerShell-heavy repo turns Claude Code and friends into a guided missile for APK analysis, CTFs, and binary diffing—if you can survive the bootstrap ritual.
To wire up public A-share, US, and HK market data into a single local dashboard and let your own AI analyze it without pretending to pick winners.
An open-source ops agent that auto-investigates alerts, correlates metrics/logs/topology, and reports root cause back to Slack or Telegram—without opening inbound ports.
PentesterFlow pairs LLMs with real pentesting tools in your terminal, demanding human approval for sensitive actions and evidence before it reports a finding.
DojoAgents autonomously reasons across global markets and your holdings to diagnose portfolio risk, track sectors, and suggest rebalancing—all from a simple screenshot upload.
It gives discretionary traders a local, two-stage LLM analyst for structured K-line data that diagnoses markets, routes strategies, and deliberately refuses to execute orders.
A set of agent skills that turns Chinese Amazon intelligence APIs into automated listing audits, keyword research, and product-selection reports for Claude and Codex users.
It exists to stop LLMs from confidently hallucinating decompiled C++ by forcing every guess through build tests and binary parity checks.








