To prevent "Value Trap" logic in Tech/Platform stocks, the Trader utilizes specific mental models distinct from traditional Value Investing.
### 1. CapEx Distinction (The "Moat" Rule)
* **Maintenance CapEx:** Spending to keep the lights on (e.g., repairing a factory). This is a **COST**.
* **Strategic CapEx:** Spending to capture a new platform (e.g., Google building Data Centers for AI). This is **DEFENSE**.
* **The Rule:** For dominant platforms, massive CapEx during a platform shift is a **BULLISH** signal of durability, not a bearish signal of inefficiency.
- `Market Analyst` now triggers `Social`, `News`, and `Fundamentals` analysts **concurrently**.
- Added `Analyst Sync` node to synchronize parallel branches.
- Added `Analyst Sync` node to synchronize parallel branches.
- reduced total runtime by ~50% by overlapping heavy LLM/Tool operations.
- **Fail Fast Scraper**: Optimized `googlenews_utils.py` to timeout after ~30s (down from 3m) when blocked, ensuring rapid failover to backup vendors.
### Fixed
- **API Error 400 (Dangling Tool Use)**: Fixed crash in `Fundamentals Analyst` and others caused by unhandled tool exceptions (e.g. Rate Limits).
- Wrapped all tools in `fundamental_data_tools.py`, `news_data_tools.py`, `core_stock_tools.py`, and `technical_indicators_tools.py` with `try/except` blocks.
- Tools now return error strings instead of crashing, ensuring stricter API compliance and system resilience.
- Removed dependency on client-side `marked.js` and CDNs.
- Reports are now fully offline-capable.
- Cleaned up JSON keys to remove `.md` extensions for cleaner data structure.
- **Google News Adapter**: Implemented `get_google_global_news` adapter in `google.py` to match the standard `(curr_date, look_back_days)` interface, adhering to the Adapter Pattern and fixing signature mismatches.
- **Robust Demo Script**: Created `run_agent.py` (replacing demo scripts) with:
- Automatic `.env` loading.
- `backend_url` handling (clearing OpenAI defaults when using Anthropic).
- Hardened configuration for "Deep Analysis" (Debate Rounds=2).
- Pre-configured Google News vendor to bypass AlphaVantage rate limits.
### Fixed
- **Rate Limit Crash**: Fixed `AlphaVantageRateLimitError` by switching default news vendor to Google in `run_agent.py`.
- **Dynamic Parameter Tuning (The Learning Loop)**: Implemented full self-reflection cycle. The Reflector agent now parses its own advice into JSON (`rsi_period`, `stop_loss_pct`), persists it to `data_cache/runtime_config.json`, and the Market Analyst loads it to tune the Regime Detector in real-time.
- **Audit Archival**: Every tuning event is now archived to `results/{TICKER}/{DATE}/runtime_config.json` for historical auditing, ensuring we can reproduce why parameters changed on any given day.
- **Atomic Persistence**: Implemented `agent_utils.write_json_atomic` to prevent race conditions during config saves.
- **Centralized Config**: Moved hardcoded paths to `default_config.py` (DRY principle).
### Fixed
- **Reflector Logic Gap**: The Reflector was previously "shouting into the void"—making suggestions but having no mechanism to apply them. This circuit is now closed.
- **Relative Strength Determinism**: Upgraded `market_analyst.py` to calculate a mathematical `risk_multiplier` (0.0x - 1.5x) based on the Asset Regime vs. SPY Regime correlation, removing LLM "confidence" hallucinations from position sizing.
- **Portfolio Awareness (Rule 72)**: Implemented State Persistence (`portfolio`, `cash_balance`) and a hard-coded Stop Loss check in `trading_graph.py`. If a position's unrealized PnL drops below -10%, the system forces a "LIQUIDATE" order, bypassing all AI debate.
- **Self-Tuning Architecture**: Updated `reflection.py` to output a structured JSON block (`UPDATE_PARAMETERS`) instead of prose advice, enabling future automated parameter optimization.
Override Logic Mismatches: Fixed critical Enum-to-String type mismatch in
apply_trend_override
that was silencing the "Safety Valve" logic.
Data Pipeline Failures: Injected robust error handling and type checking in
market_analyst.py
to identify why RegimeDetector receives invalid data (causing "UNKNOWN" regimes).
Gemini 404 Errors: Removed invalid/deprecated model names causing 404s
route_to_vendor
to treat empty/whitespace-only results as failures, ensuring the fallback chain continues to the next vendor.
Analyst Nodes: Removed destructive logic that was overwriting real company names with ticker symbols in the anonymizer state.
Fixed
News Vendors: Resolved a TypeError in Google news results and removed the non-functional OpenAI news implementations.
Report Generator: Fixed a NameError related to sys and restored path-parsing logic in
- **Configurable Embeddings Truncation**: Added `EMBEDDING_TRUNCATION_LIMIT` env var (default 1000) to prevent `413 Payload Too Large` errors with local models.
- Add .env.example file with API key placeholders
- Update README.md with .env file setup instructions
- Add dotenv loading in main.py for environment variables
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add data vendor configuration examples in README and main.py showing how to configure Alpha Vantage as the primary data provider. Update documentation to reflect the current default behavior of using Alpha Vantage for real-time market data access.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Replace hardcoded column indices with column name lookup
- Add mapping for all supported indicators to their expected CSV column names
- Handle missing columns gracefully with descriptive error messages
- Strip whitespace from header parsing for reliability
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Replace FinnHub with Alpha Vantage API in README documentation
- Implement comprehensive Alpha Vantage modules:
- Stock data (daily OHLCV with date filtering)
- Technical indicators (SMA, EMA, MACD, RSI, Bollinger Bands, ATR)
- Fundamental data (overview, balance sheet, cashflow, income statement)
- News and sentiment data with insider transactions
- Update news analyst tools to use ticker-based news search
- Integrate Alpha Vantage vendor methods into interface routing
- Maintain backward compatibility with existing vendor system
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Added support for running CLI and Ollama server via Docker
- Introduced tests for local embeddings model and standalone Docker setup
- Enabled conditional Ollama server launch via LLM_PROVIDER
swj.premkumar
Edward Sun
Edward Sun
Yijia Xiao
luohy15
mirza-samad-ahmed-baig
Max Wong
Geeta Chauhan
Huijae Lee
Zhongyi Lu