PRISM-INSIGHT
These are 13 specialized AI agents that analyze Korean and US stocks in real-time, and this is open source and free to use.
AI translation
A cinematic multi-asset terminal for the US & Indian markets: stocks, bonds, mutual funds, currencies, and derivatives with derivative-aware analytics.
FinanceTypeScript · PythonGitHub @abrarakhanOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Self-hosted India-first market intelligence and trade-management platform with Strong Swing, Momentum Ignition V2, Python ML advisory foundations, live NSE/BSE data, ICICI Breeze controls, broker reconciliation, adaptive allocation, and ROI/XIRR.
Taken from the repository README
Loading the full README.
Read the full README ↗InvestoGenie is a self-hosted market research and trade-management application for Indian and US equities. Next.js serves the web terminal and mobile APIs; PostgreSQL stores market data, user portfolios, frozen trade plans and operational history.
Code and production baseline reviewed 26 October 2026 through runtime revision 1c7aac3. See STATUS.md for the dated verification boundary.
| Route | Purpose |
|---|---|
/markets/in, /markets/us | Market overview and charts |
/terminal/[market] | India (in) or US (us) terminal |
/terminal/[market]/stocks | Fundamental and price-action Stock Screener, saved screens, natural-language queries and exports |
/terminal/[market]/screener | Buy-side Swing Candidates; India is NSE-only and capped to the top 20 |
/terminal/[market]/strong-swing | Strong Swing, legacy/V2 Momentum Ignition and Python Trend Continuation observation |
/terminal/[market]/armed-swing | Earlier pre-breakout Armed Strong Swing plans |
/terminal/[market]/news-swing | Strong Swing with source-linked news and AI assessment |
/terminal/[market]/trade-ledger | Purchases, pending buy-stops, partial fills/exits, reconciliation and returns |
/terminal/[market]/long-term | Multi-year fundamentals and investor-inspired rankings |
/terminal/[market]/probability | 21-session probability forecasts |
/terminal/[market]/adaptive-allocation | Cost-aware two-asset backtests and saved strategies |
/terminal/[market]/forward-test | Forward measurement of recorded signals |
/terminal/in/cas, /portfolio/fund-mapping | CAS import, Gmail disclosures, AMFI mapping and fund overlap |
/data/health, /admin/sync | Freshness, coverage and ingestion health |
/settings, /help | Account/provider configuration and product guidance |
Analytics run on the server. Strong Swing and Momentum purchase flows use stored execution plans; Armed Swing stores its own immutable trigger, limit, stop, target and gate snapshot. Pending buy-stops are ledger records: creating one sends no broker order and creates no position or P&L. Confirmed partial fills create separate purchased trades while preserving the unfilled remainder.
Armed setups require verified results-calendar coverage and live execution data. Missing evidence blocks actionable states. India execution quotes require BREEZE_LIVE; a fallback quote alone cannot authorize an Armed entry.
The ledger separates cumulative purchase turnover from external capital employed after recycling retained sale proceeds. ROI uses capital employed; XIRR uses dated contributions and terminal cash plus open-position value. Broker-reconciled acquisition/proceeds/P&L can replace gross values.
Breeze automatic buying is a separate, disabled-by-default, explicitly enabled path for capped NSE cash limit buys from Strong Swing and saved Adaptive Allocation strategies. It does not automatically sell or rebalance. Research projections and notification delivery do not place orders.
| Layer | Repository dependency/configuration |
|---|---|
| Web | Next.js 16.4.2, React 19.2.4, TypeScript |
| UI | Tailwind CSS 4, Lightweight Charts, Three.js |
| Database | PostgreSQL through pg |
| Mobile | Expo ~57.0.24, React Native 0.86.3, React 19.2.3 |
| Jobs | Node scheduler and Python pipelines/workers |
| Credentials | AES-256-GCM per-user encrypted secrets |
| Deployment package | macOS launchd or Ubuntu systemd/Nginx; Linux scripts support AWS/Oracle hosts |
The supplied local commands target macOS/Linux. Use a current compatible Node.js runtime, Python 3 and PostgreSQL; several package scripts expect .venv/bin/python.
cp .env.example .env.local
npm ci
python3 -m venv .venv
.venv/bin/pip install -r pipelines/requirements.txt -r workers/requirements.txt
Create the database named in DATABASE_URL. Set distinct random SESSION_SECRET, CRON_SECRET and CREDENTIAL_ENCRYPTION_KEY values. Set one nonempty NEXT_PUBLIC_APP_URL (the example currently contains two declarations, so remove or replace the blank one). For a fresh database, apply the ordered SQL migrations before starting; see local setup.
npm run deploy:check:local
npm run dev
Open http://localhost:3000. Both npm run dev and npm start run the scheduler wrapper. dev:next and start:next start only Next.js.
The wrapper coordinates market-hours quotes, EOD ingestion, history repair, fundamentals, macro, AMFI, scans, forward tests, news, email, mobile alerts, Gmail processing, results calendars and scheduled ML snapshots. Breeze supplies live India data; exchange Bhavcopy, Yahoo/Google and FRED support the other ingestion paths. News providers include Marketaux, GNews, NewsAPI and Alpha Vantage.
Momentum scans run during market hours even when an account has no registered push device. India actionability uses a separately verified current-session Breeze OHLCV/volume bar; completed history remains immutable during the session. Candidate state changes are retained for audit, while push delivery still requires an enabled mobile token.
The ML implementation is a Python feature/label foundation and lifecycle schema. The repository does not include a complete training, promotion or inference pipeline. See ML methodology.
npm test
npm run lint
npm run build
Database checks need a configured PostgreSQL instance. Linux releases use deploy/oracle/deploy-release.sh; it applies its explicit migration list from 0033 through 0059 and assumes the earlier schema already exists. After updating the checkout, it re-executes the checked-out release script once so same-release migration additions are honored. It runs dependency audit/build/schema checks, but does not run the test suite.
See architecture, capabilities, status, deployment, Stock Screener, email setup, and mobile.