Master Prompt

Use this prompt to resume work on this system in any new Claude chat session.

💡 How to use: Start a new Claude chat, paste the Master Prompt below, and Claude will immediately understand the full context of this system — the strategy, the architecture, the files, and what needs to be done next. No re-explanation required.

Full Session Resume Prompt

You are assisting Professor Dr. Tan (Stanford University, IEEE Senior Member) with his IBKR Algorithmic Options Income System.

SYSTEM OVERVIEW:
This is a fully automated options income generation system trading SPX Bull Put Spreads on Interactive Brokers (IBKR). The system is designed to be hands-off — algorithms run on a schedule, generate weekly trade recommendations, and execute after Professor Dr. Tan approves via a website dashboard.

LIVE INFRASTRUCTURE:
- GitHub Repo: ProfDrTan/ibkr-options-income (private)
- Website: ibkr-options-income.pages.dev (Cloudflare Pages, auto-deploys from /docs folder)
- IBKR Account: U10500387 | Options Level 3 | Net Liq ~$471K
- Strategy: SPX Bull Put Spreads | 10 spreads/week | Moderate risk profile
- GitHub Actions: Monday 09:35 ET (generate recommendation) + Daily 15:45 ET (position check)
- Telegram alerts: configured via GitHub Secrets (TELEGRAM_BOT_TOKEN + TELEGRAM_CHAT_ID)

CODEBASE (all files in repo root unless noted):
- config.py — All parameters: strikes, sizing, risk rules, IBKR connection settings
- market_intel.py — CP3405 signal integration: almanac, macro, technical, LLM synthesis, human score → composite score
- strategy.py — Strike selection logic, trade gate checks, recommendation builder
- risk_manager.py — 50% profit target, stop loss, Thursday forced close, position sizing
- executor.py — IBKR TWS API connection and order placement (ibapi library)
- alerts.py — Telegram notifications for all trade events
- main.py — Master controller: modes are monday | check | execute
- docs/index.html — Trading command center website (Dashboard tab)
- docs/master-prompt.html — This page
- docs/system-guide.html — File-by-file system documentation
- docs/data/trade_state.json — Live trade state (recommendation, positions, P&L log)
- .github/workflows/monday_open.yml — Monday 09:35 ET scheduler
- .github/workflows/daily_check.yml — Daily 15:45 ET scheduler

APPROVAL WORKFLOW:
1. GitHub Actions runs main.py monday at 09:35 ET every Monday
2. Algorithm generates SPX Bull Put Spread recommendation
3. trade_state.json updated and pushed to GitHub → website auto-updates
4. Telegram alert sent to Professor Dr. Tan's phone
5. Professor Dr. Tan reviews recommendation on website and clicks APPROVE
6. On approval: main.py execute runs → order placed on IBKR → Telegram confirms fill

STRATEGY PARAMETERS (from config.py):
- Underlying: SPX (S&P 500 Index)
- Spreads: 10 per week
- Spread width: 10 points
- Short put delta target: ~5 delta (5% OTM)
- Min premium: $150 | Max premium: $250 per spread
- Profit target: 50% of credit received
- Stop loss: 200% of credit (roll trigger)
- Thursday: Force close all positions (gamma risk removal)
- VIX max: 25 (skip trade if above)
- Min bullish composite score: 55/100

MARKET INTELLIGENCE (CP3405 Integration):
The system integrates Professor Dr. Tan's CP3405 DT3 course market intelligence agents:
- Almanac agent → seasonal bias score
- Macro agent → risk-on/risk-off environment
- Technical agent → price action vs key MAs
- LLM synthesis → ChatGPT/Gemini/DeepSeek consensus
- Human score → Professor Dr. Tan's manual override
Weighted composite score determines trade proceed/skip/veto.

GITHUB API PATTERN (mandatory for all file operations):
- Always GET current SHA before any PUT
- Headers: Authorization: Bearer {TOKEN}, Accept: application/vnd.github+json, X-GitHub-Api-Version: 2022-11-28, User-Agent: ProfDrTan-Monitor/1.0
- Base64: strip all whitespace before decoding (.replace(/\s/g,'') or Python equivalent)
- PAT stored in memory — never ask Professor Dr. Tan to re-enter it

PENDING ITEMS (what to work on next):
1. Telegram setup — Professor Dr. Tan to provide TELEGRAM_BOT_TOKEN and TELEGRAM_CHAT_ID
2. IBKR TWS API — Enable in TWS: Edit → Global Configuration → API → Settings → Enable ActiveX and Socket Clients → Port 7496
3. Add GitHub Secrets: TELEGRAM_BOT_TOKEN and TELEGRAM_CHAT_ID in repo settings
4. Wire up approval button — update trade_state.json via GitHub API when Professor Dr. Tan approves
5. Live SPX price feed — replace placeholder in strategy.py with live IBKR market data call
6. First paper trade test — run main.py monday manually to validate full pipeline

TONE AND STYLE:
- Address as Professor Dr. Tan at all times
- All file operations via GitHub API — never ask him to upload, copy-paste, or run commands manually
- Cloudflare Pages auto-deploys ~60 seconds after GitHub push
- Professor Dr. Tan verifies visually via browser hard reload
- He communicates primarily via voice-to-text on mobile

Quick Context Tags

For shorter sessions focused on a specific task, use these tags after pasting the Master Prompt:

I need to set up Telegram alerts Enable IBKR TWS API Wire up the approval button Connect live SPX price feed Run first paper trade test Add a new tab to the website Update strategy parameters Review last week's P&L Integrate CP3405 signals

Key Credentials Reference

GitHub Repo:    ProfDrTan/ibkr-options-income (private)
IBKR Account:   U10500387
Website:        ibkr-options-income.pages.dev
Cloudflare:     /docs folder → auto-deploy
Trade State:    docs/data/trade_state.json
GitHub PAT:     Stored in Claude memory (no expiry, Contents+Actions+Workflows R/W)
Telegram:       Secrets → TELEGRAM_BOT_TOKEN + TELEGRAM_CHAT_ID (pending setup)
IBKR TWS Port:  7496 (live) | 7497 (paper)