Releases: Hitheshkaranth/OpenTerminalUI
OpenTerminalUI 0.5.0
OpenTerminalUI 0.5.0 turns the terminal from a research surface into an autonomous research agent. The headline is a tool-using AI Research Agent — including a multi-agent debate and a new Strategy Lab that closes the loop from idea → backtest → out-of-sample validation — alongside a real RAG knowledge base and a much deeper backtesting engine. Multi-market (US NYSE/NASDAQ + India NSE/BSE) throughout.
✨ AI Research Agent
- Conversational console — a global slide-over panel (
Ctrl/Cmd + J) invokable from any screen, with a chat thread, streaming step trace, and crafted artifact renderers (snapshot cards, screener tables, equity curves). - Tool-using agentic loop — autonomously calls read-only tools (screener, full snapshot, multi-ticker compare, technicals, setup scan, backtests, robustness validation, research RAG) and reasons over the results.
- Screen-aware context — defaults to the stock and market you currently have open, so "tell me about this stock" resolves correctly.
- Multi-agent debate mode — an analyst team (fundamental / sentiment / technical) feeds a bull-vs-bear debate that a portfolio manager resolves into a
BUY / HOLD / SELLdecision with a conviction score. - 🆕 Strategy Lab — a bounded, read-only loop that proposes a strategy, backtests it, changes one variable to iterate toward a target metric, then runs mandatory out-of-sample validation (permutation + multi-window robustness) and reports an honest verdict — refusing to call a curve-fit result an edge. Flag-gated, capped on rounds and wall-clock.
- MCP server — the read-only agent tools are exposed over MCP.
- Provider-flexible & resilient — OpenRouter, OpenAI, or local LM Studio, with an automatic free-model fallback chain and per-phase model routing; degrades gracefully on rate limits and empty completions.
🔬 Research Knowledge Base (RAG)
- Full-text knowledge base with arXiv ingest, multi-source ingest (URL / PDF), and TF-IDF retrieval, surfaced via a dedicated Research page and a
search_researchagent tool.
📈 Backtesting & Quant
- Vectorized parameter-sweep engine + Param Sweep tab.
- Robustness suite — permutation test + multi-window analysis to separate real edges from curve-fitting.
- 30+ performance metrics plus scenario projections (new Metrics tab).
- Additional risk rules — TakeProfit, Cooldown, Scaling — with per-symbol cost modeling.
🐛 Notable fixes
- Security Hub financials now render (the metric×period matrix is correctly pivoted into period rows) — annual + quarterly.
- News by-ticker pointed at the correct backend route (was hitting a non-existent
/v1/news/tickerendpoint and always showing empty). - Watchlist gains a REST quote fallback so prices show when the live feed is idle;
/quotesnow falls back to Yahoo for US symbols when Finnhub is unconfigured. - FMP adapter migrated from the retired v3 API to the stable API.
- News emotion routed through the agent LLM providers (OpenRouter / LM Studio).
- Hardened
exec()sandboxes against dunder traversal (RCE prevention) and fixed several endpoints that 500'd on bad upstream/options data.
📦 Install / docs
- One-command installer with auto-seeded admin and unified config.
- README refreshed with US + India screenshots, a seeded demo portfolio, and the running-agent captures.
Full changelog: v0.4.0...v0.5.0
OpenTerminalUI 0.4.0
Highlights
- Added Pair Trading Lab documentation for cointegration screening, hedge-ratio estimation, spread z-score diagnostics, half-life analysis, and mean-reversion trade simulations.
- Updated the app release version to 0.4.0 across frontend package metadata, app constants, README, and docs site.
- Documented the expanded 16+ strategy template catalog, including Awesome Oscillator, Heikin-Ashi, Parabolic SAR, Dual Thrust, shooting star reversal, and Bollinger W/M patterns.
- Fixed journal and notification router registration for newer FastAPI versions so backend test collection no longer fails with empty prefix/path routes.
OpenTerminalUI 0.3.0
Release 0.3.0.\n\nDocker image asset:\n- openterminalui-0.3.0-linux-arm64.tar.gz\n- Image tag: openterminalui:0.3.0\n- Platform: linux/arm64\n- SHA256: 972147f196ff32721003c84071ad92cf8d546ac0a69a4715782130a633e3d005\n\nLoad locally with:\nsh\ngunzip -c openterminalui-0.3.0-linux-arm64.tar.gz | docker load\n
v0.1.1
Release 0.1.1\n\nArtifact:\n- OpenTerminalUI-0.1.1-macos-arm64.zip\n\nSHA-256:\n- c11fe9ca655af9fdac29680918b608fe85477a69fe1d5b342e8b9a4ed906be74
v1.0.0 — OpenTerminalUI Public Release
OpenTerminalUI v1.0.0 — Public Release
Analyze. Trade. Optimize.
Open-source Bloomberg-style terminal for India (NSE/BSE) and US (NYSE/NASDAQ) markets.
Highlights
- Multi-Asset Terminal — Equities, F&O derivatives, crypto, mutual funds, and fixed income in one workspace
- Advanced Charting — TradingView-style chart workstation with 40+ technical indicators, drawing tools, multi-chart sync, and volume profile overlays
- Pro Analytics — Portfolio monitoring, backtesting lab, risk dashboard, and sector rotation (RRG) analysis
- Real-Time Data — WebSocket-driven live quotes, order book depth, and market heatmaps
- Stock Screener — Custom filter expressions with real-time screening across NSE/BSE/US exchanges
- Keyboard-First UX — Command palette (Ctrl+K), workspace hotkeys, and terminal-noir dark theme
- AI Research Copilot — Natural-language equity research assistant
- Economic Data Terminal — Macro indicators and central bank data dashboard
Downloads
| Asset | Description |
|---|---|
OpenTerminalUI-Container.tar |
Pre-built Docker image — load with docker load -i OpenTerminalUI-Container.tar then docker run -p 8000:8000 openterminalui |
OpenTerminalUI-Windows-App.zip |
Standalone Windows desktop app — extract and run OpenTerminalUI.exe |
Docker Quick Start
# Load the pre-built image
docker load -i OpenTerminalUI-Container.tar
# Run the terminal
docker run -p 8000:8000 openterminalui
# Open in browser
# http://localhost:8000Windows Quick Start
- Download and extract
OpenTerminalUI-Windows-App.zip - Run
OpenTerminalUI.exe - The terminal opens automatically at
http://127.0.0.1:8000
Build from Source
# Clone
git clone https://github.com/Hitheshkaranth/OpenTerminalUI.git
cd OpenTerminalUI
# Backend
pip install -r backend/requirements.txt
# Frontend
cd frontend && npm install && npm run build && cd ..
# Run
uvicorn backend.main:app --host 0.0.0.0 --port 8000Tech Stack
Python 3.11 · FastAPI · React 18 · TypeScript · Vite 6 · Tailwind CSS · SQLAlchemy · Redis · WebSocket · Docker
Screenshots
MIT License · Built with Claude Code



