feat: integrate scenario engine and playbook into main trading loop (issue #84) #110

agentson · 2026-02-09T22:47:30+09:00

agentson commented

2026-02-09 22:47:30 +09:00

Summary

Replace reactive brain.decide() with proactive scenario_engine.evaluate() in the main trading loop
Initialize V2 components in run(): ContextSelector, ScenarioEngine, PlaybookStore, PreMarketPlanner
Enrich market_data with scanner metrics (RSI, volume_ratio) for ScenarioEngine condition matching
Build portfolio_data dict with portfolio_pnl_pct, total_cash, total_eval for global rule evaluation
Wrap ScenarioMatch in TradeDecision for backward compatibility with downstream code (order execution, logging, risk)
Add scenario match notifications via Telegram when conditions are met
Update run_daily_session() to generate playbook via PreMarketPlanner then evaluate per-stock
Update realtime loop: playbook generation on first scan per market, periodic rescan, playbook cleanup on market close
6 new integration tests in TestScenarioEngineIntegration:
- Market data enrichment with/without scanner data
- Scenario matched notification sent/not sent
- Decision logger receives scenario match details
- REDUCE_ALL action doesn't execute orders

Test Plan

All 22 tests in tests/test_main.py pass
Full test suite: 486 passed (1 pre-existing failure in test_context.py)
Ruff lint clean (only pre-existing F841 market_scanner unused variable)
Manual verification: paper trading mode with scenario engine

Closes #84

🤖 Generated with Claude Code

## Summary - **Replace reactive `brain.decide()` with proactive `scenario_engine.evaluate()`** in the main trading loop - **Initialize V2 components** in `run()`: `ContextSelector`, `ScenarioEngine`, `PlaybookStore`, `PreMarketPlanner` - **Enrich market_data** with scanner metrics (RSI, volume_ratio) for ScenarioEngine condition matching - **Build portfolio_data** dict with `portfolio_pnl_pct`, `total_cash`, `total_eval` for global rule evaluation - **Wrap ScenarioMatch in TradeDecision** for backward compatibility with downstream code (order execution, logging, risk) - **Add scenario match notifications** via Telegram when conditions are met - **Update run_daily_session()** to generate playbook via PreMarketPlanner then evaluate per-stock - **Update realtime loop**: playbook generation on first scan per market, periodic rescan, playbook cleanup on market close - **6 new integration tests** in `TestScenarioEngineIntegration`: - Market data enrichment with/without scanner data - Scenario matched notification sent/not sent - Decision logger receives scenario match details - REDUCE_ALL action doesn't execute orders ## Test Plan - [x] All 22 tests in `tests/test_main.py` pass - [x] Full test suite: 486 passed (1 pre-existing failure in `test_context.py`) - [x] Ruff lint clean (only pre-existing `F841 market_scanner` unused variable) - [ ] Manual verification: paper trading mode with scenario engine Closes #84 🤖 Generated with [Claude Code](https://claude.com/claude-code)

agentson added 2 commits 2026-02-09 22:47:30 +09:00

feat: integrate scenario engine and playbook into main trading loop (issue #84 ) 98c4a2413c

Replace brain.decide() with scenario_engine.evaluate() in trading_cycle
and brain.decide_batch() with per-stock scenario evaluation in
run_daily_session. Initialize PreMarketPlanner, ScenarioEngine, and
PlaybookStore in run(). Add pre-market playbook generation on market
open (1 Gemini call per market per day), market_data enrichment from
scanner metrics (rsi, volume_ratio), portfolio_data for global rules,
scenario match notifications, and playbook lifecycle management.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

fix: resolve lint issues in main.py and test_main.py

CI / test (pull_request) Has been cancelled

Details

b2312fbe01

Remove unused imports (sys, ScenarioMatch, asyncio, StockPlaybook),
fix import ordering, and split long lines for ruff compliance.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

agentson referenced this issue from a commit

2026-02-09 23:00:11 +09:00

fix: PR review — DB reload, market-local date, market-scoped scan_candidates

agentson added 1 commit 2026-02-09 23:00:11 +09:00

fix: PR review — DB reload, market-local date, market-scoped scan_candidates

CI / test (pull_request) Has been cancelled

Details

d64e072f06

Address PR #110 review findings:

1. High — Realtime mode now loads playbook from DB before calling Gemini,
   preventing duplicate API calls on process restart (4/day budget).
2. Medium — Pass market-local date (via market.timezone) to
   generate_playbook() and _empty_playbook() instead of date.today().
3. Medium — scan_candidates restructured from {stock_code: candidate}
   to {market_code: {stock_code: candidate}} to prevent KR/US symbol
   collision.

New test: test_scan_candidates_market_scoped verifies cross-market
isolation.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

jihoson merged commit 9171e54652 into main

2026-02-09 23:18:24 +09:00

jihoson referenced this issue from a commit

2026-02-09 23:18:24 +09:00

Merge pull request 'feat: integrate scenario engine and playbook into main trading loop (issue #84)' (#110) from feature/issue-84-main-integration into main

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: jihoson/The-Ouroboros#110