merge: sync with upstream TauricResearch/TradingAgents v0.2.2
Merges upstream/main into our fork, bringing in: - Security: Remove chainlit (CVE-2026-22218), patch LangGrinch vulnerability - Bug fixes: debate state init, UTF-8 encoding, stock data parsing, debate round config - Features: OpenAI Responses API, five-tier rating scale, Anthropic effort support, exchange-qualified tickers, yfinance retry with exponential backoff - Refactor: risk_manager renamed to portfolio_manager Conflicts resolved preserving our fork's: - Per-tier LLM config (quick/mid/deep think with provider overrides) - Separate tool files (fundamental_data_tools, core_stock_tools, etc.) - Custom tools (get_ttm_analysis, get_peer_comparison, get_sector_relative, get_macro_regime) - Dynamic Ollama model fetching - Enhanced fundamentals analyst prompt with TTM analysis - Hardened stockstats/yfinance data pipeline (_load_or_fetch_ohlcv) - AgentOS observability layer, scanner pipeline, portfolio management Tests: 727 passed, 14 skipped
This commit is contained in:
copilot-swe-agent[bot]
commit
4c186a55e8
13
README.md
13
README.md
|
|
@ -28,7 +28,7 @@
|
|||
# TradingAgents: Multi-Agents LLM Financial Trading Framework
|
||||
|
||||
## News
|
||||
- [2026-03] **TradingAgents v0.2.1** released with GPT-5.4, Gemini 3.1, Claude 4.6 model coverage and improved system stability.
|
||||
- [2026-03] **TradingAgents v0.2.2** released with GPT-5.4/Gemini 3.1/Claude 4.6 model coverage, five-tier rating scale, OpenAI Responses API, Anthropic effort control, and cross-platform stability.
|
||||
- [2026-02] **TradingAgents v0.2.0** released with multi-provider LLM support (GPT-5.x, Gemini 3.x, Claude 4.x, Grok 4.x) and improved system architecture.
|
||||
- [2026-01] **Trading-R1** [Technical Report](https://arxiv.org/abs/2509.11420) released, with [Terminal](https://github.com/TauricResearch/Trading-R1) expected to land soon.
|
||||
|
||||
|
|
@ -112,9 +112,9 @@ conda create -n tradingagents python=3.13
|
|||
conda activate tradingagents
|
||||
```
|
||||
|
||||
Install dependencies:
|
||||
Install the package and its dependencies:
|
||||
```bash
|
||||
pip install -r requirements.txt
|
||||
pip install .
|
||||
```
|
||||
|
||||
### Required APIs
|
||||
|
|
@ -139,11 +139,12 @@ cp .env.example .env
|
|||
|
||||
### CLI Usage
|
||||
|
||||
You can also try out the CLI directly by running:
|
||||
Launch the interactive CLI:
|
||||
```bash
|
||||
python -m cli.main
|
||||
tradingagents # installed command
|
||||
python -m cli.main # alternative: run directly from source
|
||||
```
|
||||
You will see a screen where you can select your desired tickers, date, LLMs, research depth, etc.
|
||||
You will see a screen where you can select your desired tickers, analysis date, LLM provider, research depth, and more.
|
||||
|
||||
<p align="center">
|
||||
<img src="assets/cli/cli_init.png" width="100%" style="display: inline-block; margin: 0 2%;">
|
||||
|
|
|
|||
23
cli/main.py
23
cli/main.py
|
|
@ -492,7 +492,7 @@ def _ask_provider_thinking_config(provider: str):
|
|||
def get_user_selections():
|
||||
"""Get all user selections before starting the analysis display."""
|
||||
# Display ASCII art welcome message
|
||||
with open("./cli/static/welcome.txt", "r", encoding="utf-8") as f:
|
||||
with open(Path(__file__).parent / "static" / "welcome.txt", "r", encoding="utf-8") as f:
|
||||
welcome_ascii = f.read()
|
||||
|
||||
# Create welcome box content
|
||||
|
|
@ -531,7 +531,9 @@ def get_user_selections():
|
|||
# Step 1: Ticker symbol
|
||||
console.print(
|
||||
create_question_box(
|
||||
"Step 1: Ticker Symbol", "Enter the ticker symbol to analyze", "SPY"
|
||||
"Step 1: Ticker Symbol",
|
||||
"Enter the exact ticker symbol to analyze, including exchange suffix when needed (examples: SPY, CNC.TO, 7203.T, 0700.HK)",
|
||||
"SPY",
|
||||
)
|
||||
)
|
||||
selected_ticker = get_ticker()
|
||||
|
|
@ -862,9 +864,11 @@ ANALYST_REPORT_MAP = {
|
|||
|
||||
|
||||
def update_analyst_statuses(message_buffer, chunk):
|
||||
"""Update all analyst statuses based on current report state.
|
||||
"""Update analyst statuses based on accumulated report state.
|
||||
|
||||
Logic:
|
||||
- Store new report content from the current chunk if present
|
||||
- Check accumulated report_sections (not just current chunk) for status
|
||||
- Analysts with reports = completed
|
||||
- First analyst without report = in_progress
|
||||
- Remaining analysts without reports = pending
|
||||
|
|
@ -879,11 +883,16 @@ def update_analyst_statuses(message_buffer, chunk):
|
|||
|
||||
agent_name = ANALYST_AGENT_NAMES[analyst_key]
|
||||
report_key = ANALYST_REPORT_MAP[analyst_key]
|
||||
has_report = bool(chunk.get(report_key))
|
||||
|
||||
# Capture new report content from current chunk
|
||||
if chunk.get(report_key):
|
||||
message_buffer.update_report_section(report_key, chunk[report_key])
|
||||
|
||||
# Determine status from accumulated sections, not just current chunk
|
||||
has_report = bool(message_buffer.report_sections.get(report_key))
|
||||
|
||||
if has_report:
|
||||
message_buffer.update_agent_status(agent_name, "completed")
|
||||
message_buffer.update_report_section(report_key, chunk[report_key])
|
||||
elif not found_active:
|
||||
message_buffer.update_agent_status(agent_name, "in_progress")
|
||||
found_active = True
|
||||
|
|
@ -1103,9 +1112,9 @@ def run_analysis():
|
|||
content = obj.report_sections[section_name]
|
||||
if content:
|
||||
file_name = f"{section_name}.md"
|
||||
text = "\n".join(str(item) for item in content) if isinstance(content, list) else content
|
||||
with open(report_dir / file_name, "w", encoding="utf-8") as f:
|
||||
f.write(content)
|
||||
|
||||
f.write(text)
|
||||
return wrapper
|
||||
|
||||
message_buffer.add_message = save_message_decorator(message_buffer, "add_message")
|
||||
|
|
|
|||
30
cli/utils.py
30
cli/utils.py
|
|
@ -9,6 +9,9 @@ from cli.models import AnalystType
|
|||
console = Console()
|
||||
|
||||
|
||||
|
||||
TICKER_INPUT_EXAMPLES = "Examples: SPY, CNC.TO, 7203.T, 0700.HK"
|
||||
|
||||
def _fetch_ollama_models(base_url: str = "http://localhost:11434") -> list[tuple[str, str]]:
|
||||
"""Fetch available models from a running Ollama instance."""
|
||||
try:
|
||||
|
|
@ -31,7 +34,7 @@ ANALYST_ORDER = [
|
|||
def get_ticker() -> str:
|
||||
"""Prompt the user to enter a ticker symbol."""
|
||||
ticker = questionary.text(
|
||||
"Enter the ticker symbol to analyze:",
|
||||
f"Enter the exact ticker symbol to analyze ({TICKER_INPUT_EXAMPLES}):",
|
||||
validate=lambda x: len(x.strip()) > 0 or "Please enter a valid ticker symbol.",
|
||||
style=questionary.Style(
|
||||
[
|
||||
|
|
@ -45,6 +48,11 @@ def get_ticker() -> str:
|
|||
console.print("\n[red]No ticker symbol provided. Exiting...[/red]")
|
||||
exit(1)
|
||||
|
||||
return normalize_ticker_symbol(ticker)
|
||||
|
||||
|
||||
def normalize_ticker_symbol(ticker: str) -> str:
|
||||
"""Normalize ticker input while preserving exchange suffixes."""
|
||||
return ticker.strip().upper()
|
||||
|
||||
|
||||
|
|
@ -410,6 +418,26 @@ def ask_openai_reasoning_effort() -> str:
|
|||
).ask()
|
||||
|
||||
|
||||
def ask_anthropic_effort() -> str | None:
|
||||
"""Ask for Anthropic effort level.
|
||||
|
||||
Controls token usage and response thoroughness on Claude 4.5+ and 4.6 models.
|
||||
"""
|
||||
return questionary.select(
|
||||
"Select Effort Level:",
|
||||
choices=[
|
||||
questionary.Choice("High (recommended)", "high"),
|
||||
questionary.Choice("Medium (balanced)", "medium"),
|
||||
questionary.Choice("Low (faster, cheaper)", "low"),
|
||||
],
|
||||
style=questionary.Style([
|
||||
("selected", "fg:cyan noinherit"),
|
||||
("highlighted", "fg:cyan noinherit"),
|
||||
("pointer", "fg:cyan noinherit"),
|
||||
]),
|
||||
).ask()
|
||||
|
||||
|
||||
def ask_gemini_thinking_config() -> str | None:
|
||||
"""Ask for Gemini thinking configuration.
|
||||
|
||||
|
|
|
|||
|
|
@ -4,14 +4,13 @@ build-backend = "setuptools.build_meta"
|
|||
|
||||
[project]
|
||||
name = "tradingagents"
|
||||
version = "0.2.1"
|
||||
version = "0.2.2"
|
||||
description = "TradingAgents: Multi-Agents LLM Financial Trading Framework"
|
||||
readme = "README.md"
|
||||
requires-python = ">=3.10"
|
||||
dependencies = [
|
||||
"langchain-core>=0.3.81",
|
||||
"backtrader>=1.9.78.123",
|
||||
"chainlit>=2.5.5",
|
||||
"langchain-anthropic>=0.3.15",
|
||||
"langchain-experimental>=0.3.4",
|
||||
"langchain-google-genai>=2.1.5",
|
||||
|
|
@ -62,3 +61,6 @@ markers = [
|
|||
"paid_tier: marks tests that require a paid Finnhub subscription (free tier returns HTTP 403)",
|
||||
]
|
||||
addopts = "--ignore=tests/integration --ignore=tests/e2e --disable-socket --allow-unix-socket -x -q"
|
||||
|
||||
[tool.setuptools.package-data]
|
||||
cli = ["static/*"]
|
||||
|
|
|
|||
|
|
@ -1,21 +1 @@
|
|||
typing-extensions
|
||||
langchain-core
|
||||
langchain-openai
|
||||
langchain-experimental
|
||||
pandas
|
||||
yfinance
|
||||
stockstats
|
||||
langgraph
|
||||
rank-bm25
|
||||
setuptools
|
||||
backtrader
|
||||
parsel
|
||||
requests
|
||||
tqdm
|
||||
pytz
|
||||
redis
|
||||
rich
|
||||
typer
|
||||
questionary
|
||||
langchain_anthropic
|
||||
langchain-google-genai
|
||||
.
|
||||
|
|
|
|||
|
|
@ -0,0 +1,18 @@
|
|||
import unittest
|
||||
|
||||
from cli.utils import normalize_ticker_symbol
|
||||
from tradingagents.agents.utils.agent_utils import build_instrument_context
|
||||
|
||||
|
||||
class TickerSymbolHandlingTests(unittest.TestCase):
|
||||
def test_normalize_ticker_symbol_preserves_exchange_suffix(self):
|
||||
self.assertEqual(normalize_ticker_symbol(" cnc.to "), "CNC.TO")
|
||||
|
||||
def test_build_instrument_context_mentions_exact_symbol(self):
|
||||
context = build_instrument_context("7203.T")
|
||||
self.assertIn("7203.T", context)
|
||||
self.assertIn("exchange suffix", context)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
unittest.main()
|
||||
|
|
@ -97,7 +97,7 @@ class TestConditionalLogicWiring:
|
|||
)
|
||||
}
|
||||
result = cl.should_continue_risk_analysis(state)
|
||||
assert result == "Risk Judge"
|
||||
assert result == "Portfolio Manager"
|
||||
|
||||
|
||||
class TestNewModulesImportable:
|
||||
|
|
|
|||
|
|
@ -142,10 +142,10 @@ class TestRiskDebateRounds2:
|
|||
assert result == "Aggressive Analyst"
|
||||
|
||||
def test_threshold_at_6(self):
|
||||
# count=6 == 3*2=6 → route to Risk Judge
|
||||
# count=6 == 3*2=6 → route to Portfolio Manager
|
||||
state = _make_risk_state(count=6, latest_speaker="Aggressive")
|
||||
result = self.cl.should_continue_risk_analysis(state)
|
||||
assert result == "Risk Judge"
|
||||
assert result == "Portfolio Manager"
|
||||
|
||||
def test_continues_at_count_5(self):
|
||||
state = _make_risk_state(count=5, latest_speaker="Aggressive")
|
||||
|
|
|
|||
|
|
@ -0,0 +1,2 @@
|
|||
import os
|
||||
os.environ.setdefault("PYTHONUTF8", "1")
|
||||
|
|
@ -15,7 +15,7 @@ from .risk_mgmt.conservative_debator import create_conservative_debator
|
|||
from .risk_mgmt.neutral_debator import create_neutral_debator
|
||||
|
||||
from .managers.research_manager import create_research_manager
|
||||
from .managers.risk_manager import create_risk_manager
|
||||
from .managers.portfolio_manager import create_portfolio_manager
|
||||
|
||||
from .trader.trader import create_trader
|
||||
|
||||
|
|
@ -33,7 +33,7 @@ __all__ = [
|
|||
"create_neutral_debator",
|
||||
"create_news_analyst",
|
||||
"create_aggressive_debator",
|
||||
"create_risk_manager",
|
||||
"create_portfolio_manager",
|
||||
"create_conservative_debator",
|
||||
"create_social_media_analyst",
|
||||
"create_trader",
|
||||
|
|
|
|||
|
|
@ -17,8 +17,7 @@ from tradingagents.dataflows.config import get_config
|
|||
def create_fundamentals_analyst(llm):
|
||||
def fundamentals_analyst_node(state):
|
||||
current_date = state["trade_date"]
|
||||
ticker = state["company_of_interest"]
|
||||
company_name = state["company_of_interest"]
|
||||
instrument_context = build_instrument_context(state["company_of_interest"])
|
||||
|
||||
tools = [
|
||||
get_ttm_analysis,
|
||||
|
|
@ -54,7 +53,7 @@ def create_fundamentals_analyst(llm):
|
|||
" If you or any other assistant has the FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** or deliverable,"
|
||||
" prefix your response with FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** so the team knows to stop."
|
||||
" You have access to the following tools: {tool_names}.\n{system_message}"
|
||||
"For your reference, the current date is {current_date}. The company we want to look at is {ticker}",
|
||||
"For your reference, the current date is {current_date}. {instrument_context}",
|
||||
),
|
||||
MessagesPlaceholder(variable_name="messages"),
|
||||
]
|
||||
|
|
@ -63,7 +62,7 @@ def create_fundamentals_analyst(llm):
|
|||
prompt = prompt.partial(system_message=system_message)
|
||||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
prompt = prompt.partial(instrument_context=instrument_context)
|
||||
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
|
||||
|
|
|
|||
|
|
@ -10,8 +10,7 @@ def create_market_analyst(llm):
|
|||
|
||||
def market_analyst_node(state):
|
||||
current_date = state["trade_date"]
|
||||
ticker = state["company_of_interest"]
|
||||
company_name = state["company_of_interest"]
|
||||
instrument_context = build_instrument_context(state["company_of_interest"])
|
||||
|
||||
tools = [
|
||||
get_macro_regime,
|
||||
|
|
@ -46,7 +45,7 @@ Volatility Indicators:
|
|||
Volume-Based Indicators:
|
||||
- vwma: VWMA: A moving average weighted by volume. Usage: Confirm trends by integrating price action with volume data. Tips: Watch for skewed results from volume spikes; use in combination with other volume analyses.
|
||||
|
||||
- Select indicators that provide diverse and complementary information. Avoid redundancy (e.g., do not select both rsi and stochrsi). Also briefly explain why they are suitable for the given market context. When you tool call, please use the exact name of the indicators provided above as they are defined parameters, otherwise your call will fail. Please make sure to call get_stock_data first to retrieve the CSV that is needed to generate indicators. Then use get_indicators with the specific indicator names. Write a very detailed and nuanced report of the trends you observe. Do not simply state the trends are mixed, provide detailed and finegrained analysis and insights that may help traders make decisions."""
|
||||
- Select indicators that provide diverse and complementary information. Avoid redundancy (e.g., do not select both rsi and stochrsi). Also briefly explain why they are suitable for the given market context. When you tool call, please use the exact name of the indicators provided above as they are defined parameters, otherwise your call will fail. Please make sure to call get_stock_data first to retrieve the CSV that is needed to generate indicators. Then use get_indicators with the specific indicator names. Write a very detailed and nuanced report of the trends you observe. Provide specific, actionable insights with supporting evidence to help traders make informed decisions."""
|
||||
+ """ Make sure to append a Markdown table at the end of the report to organize key points in the report, organized and easy to read."""
|
||||
)
|
||||
|
||||
|
|
@ -61,7 +60,7 @@ Volume-Based Indicators:
|
|||
" If you or any other assistant has the FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** or deliverable,"
|
||||
" prefix your response with FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** so the team knows to stop."
|
||||
" You have access to the following tools: {tool_names}.\n{system_message}"
|
||||
"For your reference, the current date is {current_date}. The company we want to look at is {ticker}",
|
||||
"For your reference, the current date is {current_date}. {instrument_context}",
|
||||
),
|
||||
MessagesPlaceholder(variable_name="messages"),
|
||||
]
|
||||
|
|
@ -70,7 +69,7 @@ Volume-Based Indicators:
|
|||
prompt = prompt.partial(system_message=system_message)
|
||||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
prompt = prompt.partial(instrument_context=instrument_context)
|
||||
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
|
||||
|
|
|
|||
|
|
@ -7,7 +7,7 @@ from tradingagents.dataflows.config import get_config
|
|||
def create_news_analyst(llm):
|
||||
def news_analyst_node(state):
|
||||
current_date = state["trade_date"]
|
||||
ticker = state["company_of_interest"]
|
||||
instrument_context = build_instrument_context(state["company_of_interest"])
|
||||
|
||||
tools = [
|
||||
get_news,
|
||||
|
|
@ -15,7 +15,7 @@ def create_news_analyst(llm):
|
|||
]
|
||||
|
||||
system_message = (
|
||||
"You are a news researcher tasked with analyzing recent news and trends over the past week. Please write a comprehensive report of the current state of the world that is relevant for trading and macroeconomics. Use the available tools: get_news(query, start_date, end_date) for company-specific or targeted news searches, and get_global_news(curr_date, look_back_days, limit) for broader macroeconomic news. Do not simply state the trends are mixed, provide detailed and finegrained analysis and insights that may help traders make decisions."
|
||||
"You are a news researcher tasked with analyzing recent news and trends over the past week. Please write a comprehensive report of the current state of the world that is relevant for trading and macroeconomics. Use the available tools: get_news(query, start_date, end_date) for company-specific or targeted news searches, and get_global_news(curr_date, look_back_days, limit) for broader macroeconomic news. Provide specific, actionable insights with supporting evidence to help traders make informed decisions."
|
||||
+ """ Make sure to append a Markdown table at the end of the report to organize key points in the report, organized and easy to read."""
|
||||
)
|
||||
|
||||
|
|
@ -30,7 +30,7 @@ def create_news_analyst(llm):
|
|||
" If you or any other assistant has the FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** or deliverable,"
|
||||
" prefix your response with FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** so the team knows to stop."
|
||||
" You have access to the following tools: {tool_names}.\n{system_message}"
|
||||
"For your reference, the current date is {current_date}. We are looking at the company {ticker}",
|
||||
"For your reference, the current date is {current_date}. {instrument_context}",
|
||||
),
|
||||
MessagesPlaceholder(variable_name="messages"),
|
||||
]
|
||||
|
|
@ -39,7 +39,7 @@ def create_news_analyst(llm):
|
|||
prompt = prompt.partial(system_message=system_message)
|
||||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
prompt = prompt.partial(instrument_context=instrument_context)
|
||||
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
result = chain.invoke(state["messages"])
|
||||
|
|
|
|||
|
|
@ -8,16 +8,15 @@ from tradingagents.dataflows.config import get_config
|
|||
def create_social_media_analyst(llm):
|
||||
def social_media_analyst_node(state):
|
||||
current_date = state["trade_date"]
|
||||
ticker = state["company_of_interest"]
|
||||
company_name = state["company_of_interest"]
|
||||
instrument_context = build_instrument_context(state["company_of_interest"])
|
||||
|
||||
tools = [
|
||||
get_news,
|
||||
]
|
||||
|
||||
system_message = (
|
||||
"You are a social media and company specific news researcher/analyst tasked with analyzing social media posts, recent company news, and public sentiment for a specific company over the past week. You will be given a company's name your objective is to write a comprehensive long report detailing your analysis, insights, and implications for traders and investors on this company's current state after looking at social media and what people are saying about that company, analyzing sentiment data of what people feel each day about the company, and looking at recent company news. Use the get_news(query, start_date, end_date) tool to search for company-specific news and social media discussions. Try to look at all sources possible from social media to sentiment to news. Do not simply state the trends are mixed, provide detailed and finegrained analysis and insights that may help traders make decisions."
|
||||
+ """ Make sure to append a Markdown table at the end of the report to organize key points in the report, organized and easy to read.""",
|
||||
"You are a social media and company specific news researcher/analyst tasked with analyzing social media posts, recent company news, and public sentiment for a specific company over the past week. You will be given a company's name your objective is to write a comprehensive long report detailing your analysis, insights, and implications for traders and investors on this company's current state after looking at social media and what people are saying about that company, analyzing sentiment data of what people feel each day about the company, and looking at recent company news. Use the get_news(query, start_date, end_date) tool to search for company-specific news and social media discussions. Try to look at all sources possible from social media to sentiment to news. Provide specific, actionable insights with supporting evidence to help traders make informed decisions."
|
||||
+ """ Make sure to append a Markdown table at the end of the report to organize key points in the report, organized and easy to read."""
|
||||
)
|
||||
|
||||
prompt = ChatPromptTemplate.from_messages(
|
||||
|
|
@ -31,7 +30,7 @@ def create_social_media_analyst(llm):
|
|||
" If you or any other assistant has the FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** or deliverable,"
|
||||
" prefix your response with FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL** so the team knows to stop."
|
||||
" You have access to the following tools: {tool_names}.\n{system_message}"
|
||||
"For your reference, the current date is {current_date}. The current company we want to analyze is {ticker}",
|
||||
"For your reference, the current date is {current_date}. {instrument_context}",
|
||||
),
|
||||
MessagesPlaceholder(variable_name="messages"),
|
||||
]
|
||||
|
|
@ -40,7 +39,7 @@ def create_social_media_analyst(llm):
|
|||
prompt = prompt.partial(system_message=system_message)
|
||||
prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
|
||||
prompt = prompt.partial(current_date=current_date)
|
||||
prompt = prompt.partial(ticker=ticker)
|
||||
prompt = prompt.partial(instrument_context=instrument_context)
|
||||
|
||||
chain = prompt | llm.bind_tools(tools)
|
||||
|
||||
|
|
|
|||
|
|
@ -1,5 +1,10 @@
|
|||
def create_risk_manager(llm, memory):
|
||||
def risk_manager_node(state) -> dict:
|
||||
from tradingagents.agents.utils.agent_utils import build_instrument_context
|
||||
|
||||
|
||||
def create_portfolio_manager(llm, memory):
|
||||
def portfolio_manager_node(state) -> dict:
|
||||
|
||||
instrument_context = build_instrument_context(state["company_of_interest"])
|
||||
|
||||
history = state["risk_debate_state"]["history"]
|
||||
risk_debate_state = state["risk_debate_state"]
|
||||
|
|
@ -20,27 +25,37 @@ def create_risk_manager(llm, memory):
|
|||
|
||||
macro_context = f"\n\nCurrent Macro Regime:\n{macro_regime_report}\nEnsure your risk assessment reflects the macro environment — in risk-off regimes, apply higher standards for position entry and tighter risk controls.\n" if macro_regime_report else ""
|
||||
|
||||
prompt = f"""As the Risk Management Judge and Debate Facilitator, your goal is to evaluate the debate between three risk analysts—Aggressive, Neutral, and Conservative—and determine the best course of action for the trader. Your decision must result in a clear recommendation: Buy, Sell, or Hold. Choose Hold only if strongly justified by specific arguments, not as a fallback when all sides seem valid. Strive for clarity and decisiveness.
|
||||
prompt = f"""As the Portfolio Manager, synthesize the risk analysts' debate and deliver the final trading decision.
|
||||
{macro_context}
|
||||
|
||||
Guidelines for Decision-Making:
|
||||
1. **Summarize Key Arguments**: Extract the strongest points from each analyst, focusing on relevance to the context.
|
||||
2. **Provide Rationale**: Support your recommendation with direct quotes and counterarguments from the debate.
|
||||
3. **Refine the Trader's Plan**: Start with the trader's original plan, **{trader_plan}**, and adjust it based on the analysts' insights.
|
||||
4. **Learn from Past Mistakes**: Use lessons from **{past_memory_str}** to address prior misjudgments and improve the decision you are making now to make sure you don't make a wrong BUY/SELL/HOLD call that loses money.
|
||||
|
||||
Deliverables:
|
||||
- A clear and actionable recommendation: Buy, Sell, or Hold.
|
||||
- Detailed reasoning anchored in the debate and past reflections.
|
||||
{instrument_context}
|
||||
|
||||
---
|
||||
|
||||
**Analysts Debate History:**
|
||||
**Rating Scale** (use exactly one):
|
||||
- **Buy**: Strong conviction to enter or add to position
|
||||
- **Overweight**: Favorable outlook, gradually increase exposure
|
||||
- **Hold**: Maintain current position, no action needed
|
||||
- **Underweight**: Reduce exposure, take partial profits
|
||||
- **Sell**: Exit position or avoid entry
|
||||
|
||||
**Context:**
|
||||
- Trader's proposed plan: **{trader_plan}**
|
||||
- Lessons from past decisions: **{past_memory_str}**
|
||||
|
||||
**Required Output Structure:**
|
||||
1. **Rating**: State one of Buy / Overweight / Hold / Underweight / Sell.
|
||||
2. **Executive Summary**: A concise action plan covering entry strategy, position sizing, key risk levels, and time horizon.
|
||||
3. **Investment Thesis**: Detailed reasoning anchored in the analysts' debate and past reflections.
|
||||
|
||||
---
|
||||
|
||||
**Risk Analysts Debate History:**
|
||||
{history}
|
||||
|
||||
---
|
||||
|
||||
Focus on actionable insights and continuous improvement. Build on past lessons, critically evaluate all perspectives, and ensure each decision advances better outcomes."""
|
||||
Be decisive and ground every conclusion in specific evidence from the analysts."""
|
||||
|
||||
response = llm.invoke(prompt)
|
||||
|
||||
|
|
@ -62,4 +77,4 @@ Focus on actionable insights and continuous improvement. Build on past lessons,
|
|||
"final_trade_decision": response.content,
|
||||
}
|
||||
|
||||
return risk_manager_node
|
||||
return portfolio_manager_node
|
||||
|
|
@ -1,8 +1,11 @@
|
|||
import time
|
||||
|
||||
from tradingagents.agents.utils.agent_utils import build_instrument_context
|
||||
|
||||
|
||||
def create_research_manager(llm, memory):
|
||||
def research_manager_node(state) -> dict:
|
||||
instrument_context = build_instrument_context(state["company_of_interest"])
|
||||
history = state["investment_debate_state"].get("history", "")
|
||||
market_research_report = state["market_report"]
|
||||
sentiment_report = state["sentiment_report"]
|
||||
|
|
@ -37,6 +40,8 @@ Take into account your past mistakes on similar situations. Use these insights t
|
|||
Here are your past reflections on mistakes:
|
||||
\"{past_memory_str}\"
|
||||
|
||||
{instrument_context}
|
||||
|
||||
Here is the debate:
|
||||
Debate History:
|
||||
{history}"""
|
||||
|
|
|
|||
|
|
@ -28,7 +28,7 @@ Market Research Report: {market_research_report}
|
|||
Social Media Sentiment Report: {sentiment_report}
|
||||
Latest World Affairs Report: {news_report}
|
||||
Company Fundamentals Report: {fundamentals_report}
|
||||
Here is the current conversation history: {history} Here are the last arguments from the conservative analyst: {current_conservative_response} Here are the last arguments from the neutral analyst: {current_neutral_response}. If there are no responses from the other viewpoints, do not hallucinate and just present your point.
|
||||
Here is the current conversation history: {history} Here are the last arguments from the conservative analyst: {current_conservative_response} Here are the last arguments from the neutral analyst: {current_neutral_response}. If there are no responses from the other viewpoints yet, present your own argument based on the available data.
|
||||
|
||||
Engage actively by addressing any specific concerns raised, refuting the weaknesses in their logic, and asserting the benefits of risk-taking to outpace market norms. Maintain a focus on debating and persuading, not just presenting data. Challenge each counterpoint to underscore why a high-risk approach is optimal. Output conversationally as if you are speaking without any special formatting."""
|
||||
|
||||
|
|
|
|||
|
|
@ -29,7 +29,7 @@ Market Research Report: {market_research_report}
|
|||
Social Media Sentiment Report: {sentiment_report}
|
||||
Latest World Affairs Report: {news_report}
|
||||
Company Fundamentals Report: {fundamentals_report}
|
||||
Here is the current conversation history: {history} Here is the last response from the aggressive analyst: {current_aggressive_response} Here is the last response from the neutral analyst: {current_neutral_response}. If there are no responses from the other viewpoints, do not hallucinate and just present your point.
|
||||
Here is the current conversation history: {history} Here is the last response from the aggressive analyst: {current_aggressive_response} Here is the last response from the neutral analyst: {current_neutral_response}. If there are no responses from the other viewpoints yet, present your own argument based on the available data.
|
||||
|
||||
Engage by questioning their optimism and emphasizing the potential downsides they may have overlooked. Address each of their counterpoints to showcase why a conservative stance is ultimately the safest path for the firm's assets. Focus on debating and critiquing their arguments to demonstrate the strength of a low-risk strategy over their approaches. Output conversationally as if you are speaking without any special formatting."""
|
||||
|
||||
|
|
|
|||
|
|
@ -28,7 +28,7 @@ Market Research Report: {market_research_report}
|
|||
Social Media Sentiment Report: {sentiment_report}
|
||||
Latest World Affairs Report: {news_report}
|
||||
Company Fundamentals Report: {fundamentals_report}
|
||||
Here is the current conversation history: {history} Here is the last response from the aggressive analyst: {current_aggressive_response} Here is the last response from the conservative analyst: {current_conservative_response}. If there are no responses from the other viewpoints, do not hallucinate and just present your point.
|
||||
Here is the current conversation history: {history} Here is the last response from the aggressive analyst: {current_aggressive_response} Here is the last response from the conservative analyst: {current_conservative_response}. If there are no responses from the other viewpoints yet, present your own argument based on the available data.
|
||||
|
||||
Engage actively by analyzing both sides critically, addressing weaknesses in the aggressive and conservative arguments to advocate for a more balanced approach. Challenge each of their points to illustrate why a moderate risk strategy might offer the best of both worlds, providing growth potential while safeguarding against extreme volatility. Focus on debating rather than simply presenting data, aiming to show that a balanced view can lead to the most reliable outcomes. Output conversationally as if you are speaking without any special formatting."""
|
||||
|
||||
|
|
|
|||
|
|
@ -2,10 +2,13 @@ import functools
|
|||
import time
|
||||
import json
|
||||
|
||||
from tradingagents.agents.utils.agent_utils import build_instrument_context
|
||||
|
||||
|
||||
def create_trader(llm, memory):
|
||||
def trader_node(state, name):
|
||||
company_name = state["company_of_interest"]
|
||||
instrument_context = build_instrument_context(company_name)
|
||||
investment_plan = state["investment_plan"]
|
||||
market_research_report = state["market_report"]
|
||||
sentiment_report = state["sentiment_report"]
|
||||
|
|
@ -24,13 +27,13 @@ def create_trader(llm, memory):
|
|||
|
||||
context = {
|
||||
"role": "user",
|
||||
"content": f"Based on a comprehensive analysis by a team of analysts, here is an investment plan tailored for {company_name}. This plan incorporates insights from current technical market trends, macroeconomic indicators, and social media sentiment. Use this plan as a foundation for evaluating your next trading decision.\n\nProposed Investment Plan: {investment_plan}\n\nLeverage these insights to make an informed and strategic decision.",
|
||||
"content": f"Based on a comprehensive analysis by a team of analysts, here is an investment plan tailored for {company_name}. {instrument_context} This plan incorporates insights from current technical market trends, macroeconomic indicators, and social media sentiment. Use this plan as a foundation for evaluating your next trading decision.\n\nProposed Investment Plan: {investment_plan}\n\nLeverage these insights to make an informed and strategic decision.",
|
||||
}
|
||||
|
||||
messages = [
|
||||
{
|
||||
"role": "system",
|
||||
"content": f"""You are a trading agent analyzing market data to make investment decisions. Based on your analysis, provide a specific recommendation to buy, sell, or hold. End with a firm decision and always conclude your response with 'FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL**' to confirm your recommendation. Do not forget to utilize lessons from past decisions to learn from your mistakes. Here is some reflections from similar situatiosn you traded in and the lessons learned: {past_memory_str}""",
|
||||
"content": f"""You are a trading agent analyzing market data to make investment decisions. Based on your analysis, provide a specific recommendation to buy, sell, or hold. End with a firm decision and always conclude your response with 'FINAL TRANSACTION PROPOSAL: **BUY/HOLD/SELL**' to confirm your recommendation. Apply lessons from past decisions to strengthen your analysis. Here are reflections from similar situations you traded in and the lessons learned: {past_memory_str}""",
|
||||
},
|
||||
context,
|
||||
]
|
||||
|
|
|
|||
|
|
@ -1,6 +1,15 @@
|
|||
from langchain_core.messages import HumanMessage, RemoveMessage
|
||||
|
||||
|
||||
|
||||
def build_instrument_context(ticker: str) -> str:
|
||||
"""Describe the exact instrument so agents preserve exchange-qualified tickers."""
|
||||
return (
|
||||
f"The instrument to analyze is `{ticker}`. "
|
||||
"Use this exact ticker in every tool call, report, and recommendation, "
|
||||
"preserving any exchange suffix (e.g. `.TO`, `.L`, `.HK`, `.T`)."
|
||||
)
|
||||
|
||||
def create_msg_delete():
|
||||
def delete_messages(state):
|
||||
"""Clear messages and add placeholder for Anthropic compatibility"""
|
||||
|
|
|
|||
|
|
@ -1,6 +1,9 @@
|
|||
import time
|
||||
import logging
|
||||
|
||||
import pandas as pd
|
||||
import yfinance as yf
|
||||
from yfinance.exceptions import YFRateLimitError
|
||||
from stockstats import wrap
|
||||
from typing import Annotated
|
||||
import os
|
||||
|
|
@ -120,6 +123,25 @@ def _load_or_fetch_ohlcv(symbol: str) -> pd.DataFrame:
|
|||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def yf_retry(func, max_retries=3, base_delay=2.0):
|
||||
"""Execute a yfinance call with exponential backoff on rate limits.
|
||||
|
||||
yfinance raises YFRateLimitError on HTTP 429 responses but does not
|
||||
retry them internally. This wrapper adds retry logic specifically
|
||||
for rate limits. Other exceptions propagate immediately.
|
||||
"""
|
||||
for attempt in range(max_retries + 1):
|
||||
try:
|
||||
return func()
|
||||
except YFRateLimitError:
|
||||
if attempt < max_retries:
|
||||
delay = base_delay * (2 ** attempt)
|
||||
logger.warning(f"Yahoo Finance rate limited, retrying in {delay:.0f}s (attempt {attempt + 1}/{max_retries})")
|
||||
time.sleep(delay)
|
||||
else:
|
||||
raise
|
||||
|
||||
|
||||
class StockstatsUtils:
|
||||
@staticmethod
|
||||
def get_stock_stats(
|
||||
|
|
|
|||
|
|
@ -5,7 +5,7 @@ from dateutil.relativedelta import relativedelta
|
|||
import pandas as pd
|
||||
import yfinance as yf
|
||||
import os
|
||||
from .stockstats_utils import StockstatsUtils, YFinanceError, _clean_dataframe, _load_or_fetch_ohlcv
|
||||
from .stockstats_utils import StockstatsUtils, YFinanceError, _clean_dataframe, _load_or_fetch_ohlcv, yf_retry
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
|
|
@ -23,7 +23,7 @@ def get_YFin_data_online(
|
|||
ticker = yf.Ticker(symbol.upper())
|
||||
|
||||
# Fetch historical data for the specified date range
|
||||
data = ticker.history(start=start_date, end=end_date)
|
||||
data = yf_retry(lambda: ticker.history(start=start_date, end=end_date))
|
||||
|
||||
# Check if data is empty
|
||||
if data.empty:
|
||||
|
|
@ -247,7 +247,7 @@ def get_fundamentals(
|
|||
"""Get company fundamentals overview from yfinance."""
|
||||
try:
|
||||
ticker_obj = yf.Ticker(ticker.upper())
|
||||
info = ticker_obj.info
|
||||
info = yf_retry(lambda: ticker_obj.info)
|
||||
|
||||
if not info:
|
||||
return f"No fundamentals data found for symbol '{ticker}'"
|
||||
|
|
@ -307,9 +307,9 @@ def get_balance_sheet(
|
|||
ticker_obj = yf.Ticker(ticker.upper())
|
||||
|
||||
if freq.lower() == "quarterly":
|
||||
data = ticker_obj.quarterly_balance_sheet
|
||||
data = yf_retry(lambda: ticker_obj.quarterly_balance_sheet)
|
||||
else:
|
||||
data = ticker_obj.balance_sheet
|
||||
data = yf_retry(lambda: ticker_obj.balance_sheet)
|
||||
|
||||
if data.empty:
|
||||
return f"No balance sheet data found for symbol '{ticker}'"
|
||||
|
|
@ -337,9 +337,9 @@ def get_cashflow(
|
|||
ticker_obj = yf.Ticker(ticker.upper())
|
||||
|
||||
if freq.lower() == "quarterly":
|
||||
data = ticker_obj.quarterly_cashflow
|
||||
data = yf_retry(lambda: ticker_obj.quarterly_cashflow)
|
||||
else:
|
||||
data = ticker_obj.cashflow
|
||||
data = yf_retry(lambda: ticker_obj.cashflow)
|
||||
|
||||
if data.empty:
|
||||
return f"No cash flow data found for symbol '{ticker}'"
|
||||
|
|
@ -367,9 +367,9 @@ def get_income_statement(
|
|||
ticker_obj = yf.Ticker(ticker.upper())
|
||||
|
||||
if freq.lower() == "quarterly":
|
||||
data = ticker_obj.quarterly_income_stmt
|
||||
data = yf_retry(lambda: ticker_obj.quarterly_income_stmt)
|
||||
else:
|
||||
data = ticker_obj.income_stmt
|
||||
data = yf_retry(lambda: ticker_obj.income_stmt)
|
||||
|
||||
if data.empty:
|
||||
return f"No income statement data found for symbol '{ticker}'"
|
||||
|
|
@ -393,7 +393,7 @@ def get_insider_transactions(
|
|||
"""Get insider transactions data from yfinance."""
|
||||
try:
|
||||
ticker_obj = yf.Ticker(ticker.upper())
|
||||
data = ticker_obj.insider_transactions
|
||||
data = yf_retry(lambda: ticker_obj.insider_transactions)
|
||||
|
||||
if data is None or data.empty:
|
||||
return f"No insider transactions data found for symbol '{ticker}'"
|
||||
|
|
|
|||
|
|
@ -73,6 +73,7 @@ DEFAULT_CONFIG = {
|
|||
# Provider-specific thinking configuration (applies to all roles unless overridden)
|
||||
"google_thinking_level": _env("GOOGLE_THINKING_LEVEL"), # "high", "minimal", etc.
|
||||
"openai_reasoning_effort": _env("OPENAI_REASONING_EFFORT"), # "medium", "high", "low"
|
||||
"anthropic_effort": _env("ANTHROPIC_EFFORT"), # "high", "medium", "low"
|
||||
# Per-role provider-specific thinking configuration
|
||||
"deep_think_google_thinking_level": _env("DEEP_THINK_GOOGLE_THINKING_LEVEL"),
|
||||
"deep_think_openai_reasoning_effort": _env("DEEP_THINK_OPENAI_REASONING_EFFORT"),
|
||||
|
|
|
|||
|
|
@ -59,7 +59,7 @@ class ConditionalLogic:
|
|||
if (
|
||||
state["risk_debate_state"]["count"] >= 3 * self.max_risk_discuss_rounds
|
||||
): # 3 rounds of back-and-forth between 3 agents
|
||||
return "Risk Judge"
|
||||
return "Portfolio Manager"
|
||||
if state["risk_debate_state"]["latest_speaker"].startswith("Aggressive"):
|
||||
return "Conservative Analyst"
|
||||
if state["risk_debate_state"]["latest_speaker"].startswith("Conservative"):
|
||||
|
|
|
|||
|
|
@ -110,12 +110,12 @@ Adhere strictly to these instructions, and ensure your output is detailed, accur
|
|||
)
|
||||
invest_judge_memory.add_situations([(situation, result)])
|
||||
|
||||
def reflect_risk_manager(self, current_state, returns_losses, risk_manager_memory):
|
||||
"""Reflect on risk manager's decision and update memory."""
|
||||
def reflect_portfolio_manager(self, current_state, returns_losses, portfolio_manager_memory):
|
||||
"""Reflect on portfolio manager's decision and update memory."""
|
||||
situation = self._extract_current_situation(current_state)
|
||||
judge_decision = current_state["risk_debate_state"]["judge_decision"]
|
||||
|
||||
result = self._reflect_on_component(
|
||||
"RISK JUDGE", judge_decision, situation, returns_losses
|
||||
"PORTFOLIO MANAGER", judge_decision, situation, returns_losses
|
||||
)
|
||||
risk_manager_memory.add_situations([(situation, result)])
|
||||
portfolio_manager_memory.add_situations([(situation, result)])
|
||||
|
|
|
|||
|
|
@ -24,7 +24,7 @@ class GraphSetup:
|
|||
bear_memory,
|
||||
trader_memory,
|
||||
invest_judge_memory,
|
||||
risk_manager_memory,
|
||||
portfolio_manager_memory,
|
||||
conditional_logic: ConditionalLogic,
|
||||
):
|
||||
"""Initialize with required components."""
|
||||
|
|
@ -36,7 +36,7 @@ class GraphSetup:
|
|||
self.bear_memory = bear_memory
|
||||
self.trader_memory = trader_memory
|
||||
self.invest_judge_memory = invest_judge_memory
|
||||
self.risk_manager_memory = risk_manager_memory
|
||||
self.portfolio_manager_memory = portfolio_manager_memory
|
||||
self.conditional_logic = conditional_logic
|
||||
|
||||
def setup_graph(
|
||||
|
|
@ -103,8 +103,8 @@ class GraphSetup:
|
|||
aggressive_analyst = create_aggressive_debator(self.quick_thinking_llm)
|
||||
neutral_analyst = create_neutral_debator(self.quick_thinking_llm)
|
||||
conservative_analyst = create_conservative_debator(self.quick_thinking_llm)
|
||||
risk_manager_node = create_risk_manager(
|
||||
self.deep_thinking_llm, self.risk_manager_memory
|
||||
portfolio_manager_node = create_portfolio_manager(
|
||||
self.deep_thinking_llm, self.portfolio_manager_memory
|
||||
)
|
||||
|
||||
# Create workflow
|
||||
|
|
@ -126,7 +126,7 @@ class GraphSetup:
|
|||
workflow.add_node("Aggressive Analyst", aggressive_analyst)
|
||||
workflow.add_node("Neutral Analyst", neutral_analyst)
|
||||
workflow.add_node("Conservative Analyst", conservative_analyst)
|
||||
workflow.add_node("Risk Judge", risk_manager_node)
|
||||
workflow.add_node("Portfolio Manager", portfolio_manager_node)
|
||||
|
||||
# Define edges
|
||||
# Start with the first analyst
|
||||
|
|
@ -178,7 +178,7 @@ class GraphSetup:
|
|||
self.conditional_logic.should_continue_risk_analysis,
|
||||
{
|
||||
"Conservative Analyst": "Conservative Analyst",
|
||||
"Risk Judge": "Risk Judge",
|
||||
"Portfolio Manager": "Portfolio Manager",
|
||||
},
|
||||
)
|
||||
workflow.add_conditional_edges(
|
||||
|
|
@ -186,7 +186,7 @@ class GraphSetup:
|
|||
self.conditional_logic.should_continue_risk_analysis,
|
||||
{
|
||||
"Neutral Analyst": "Neutral Analyst",
|
||||
"Risk Judge": "Risk Judge",
|
||||
"Portfolio Manager": "Portfolio Manager",
|
||||
},
|
||||
)
|
||||
workflow.add_conditional_edges(
|
||||
|
|
@ -194,11 +194,11 @@ class GraphSetup:
|
|||
self.conditional_logic.should_continue_risk_analysis,
|
||||
{
|
||||
"Aggressive Analyst": "Aggressive Analyst",
|
||||
"Risk Judge": "Risk Judge",
|
||||
"Portfolio Manager": "Portfolio Manager",
|
||||
},
|
||||
)
|
||||
|
||||
workflow.add_edge("Risk Judge", END)
|
||||
workflow.add_edge("Portfolio Manager", END)
|
||||
|
||||
# Compile and return
|
||||
return workflow.compile()
|
||||
|
|
|
|||
|
|
@ -18,12 +18,14 @@ class SignalProcessor:
|
|||
full_signal: Complete trading signal text
|
||||
|
||||
Returns:
|
||||
Extracted decision (BUY, SELL, or HOLD)
|
||||
Extracted rating (BUY, OVERWEIGHT, HOLD, UNDERWEIGHT, or SELL)
|
||||
"""
|
||||
messages = [
|
||||
(
|
||||
"system",
|
||||
"You are an efficient assistant designed to analyze paragraphs or financial reports provided by a group of analysts. Your task is to extract the investment decision: SELL, BUY, or HOLD. Provide only the extracted decision (SELL, BUY, or HOLD) as your output, without adding any additional text or information.",
|
||||
"You are an efficient assistant that extracts the trading decision from analyst reports. "
|
||||
"Extract the rating as exactly one of: BUY, OVERWEIGHT, HOLD, UNDERWEIGHT, SELL. "
|
||||
"Output only the single rating word, nothing else.",
|
||||
),
|
||||
("human", full_signal),
|
||||
]
|
||||
|
|
|
|||
|
|
@ -136,7 +136,7 @@ class TradingAgentsGraph:
|
|||
self.bear_memory = FinancialSituationMemory("bear_memory", self.config)
|
||||
self.trader_memory = FinancialSituationMemory("trader_memory", self.config)
|
||||
self.invest_judge_memory = FinancialSituationMemory("invest_judge_memory", self.config)
|
||||
self.risk_manager_memory = FinancialSituationMemory("risk_manager_memory", self.config)
|
||||
self.portfolio_manager_memory = FinancialSituationMemory("portfolio_manager_memory", self.config)
|
||||
|
||||
# Create tool nodes
|
||||
self.tool_nodes = self._create_tool_nodes()
|
||||
|
|
@ -155,7 +155,7 @@ class TradingAgentsGraph:
|
|||
self.bear_memory,
|
||||
self.trader_memory,
|
||||
self.invest_judge_memory,
|
||||
self.risk_manager_memory,
|
||||
self.portfolio_manager_memory,
|
||||
self.conditional_logic,
|
||||
)
|
||||
|
||||
|
|
@ -201,6 +201,11 @@ class TradingAgentsGraph:
|
|||
if reasoning_effort:
|
||||
kwargs["reasoning_effort"] = reasoning_effort
|
||||
|
||||
elif provider == "anthropic":
|
||||
effort = self.config.get("anthropic_effort")
|
||||
if effort:
|
||||
kwargs["effort"] = effort
|
||||
|
||||
return kwargs
|
||||
|
||||
def _create_tool_nodes(self) -> Dict[str, ToolNode]:
|
||||
|
|
@ -341,8 +346,8 @@ class TradingAgentsGraph:
|
|||
self.reflector.reflect_invest_judge(
|
||||
self.curr_state, returns_losses, self.invest_judge_memory
|
||||
)
|
||||
self.reflector.reflect_risk_manager(
|
||||
self.curr_state, returns_losses, self.risk_manager_memory
|
||||
self.reflector.reflect_portfolio_manager(
|
||||
self.curr_state, returns_losses, self.portfolio_manager_memory
|
||||
)
|
||||
|
||||
def process_signal(self, full_signal):
|
||||
|
|
|
|||
|
|
@ -2,9 +2,26 @@ from typing import Any, Optional
|
|||
|
||||
from langchain_anthropic import ChatAnthropic
|
||||
|
||||
from .base_client import BaseLLMClient
|
||||
from .base_client import BaseLLMClient, normalize_content
|
||||
from .validators import validate_model
|
||||
|
||||
_PASSTHROUGH_KWARGS = (
|
||||
"timeout", "max_retries", "api_key", "max_tokens",
|
||||
"callbacks", "http_client", "http_async_client", "effort",
|
||||
)
|
||||
|
||||
|
||||
class NormalizedChatAnthropic(ChatAnthropic):
|
||||
"""ChatAnthropic with normalized content output.
|
||||
|
||||
Claude models with extended thinking or tool use return content as a
|
||||
list of typed blocks. This normalizes to string for consistent
|
||||
downstream handling.
|
||||
"""
|
||||
|
||||
def invoke(self, input, config=None, **kwargs):
|
||||
return normalize_content(super().invoke(input, config, **kwargs))
|
||||
|
||||
|
||||
class AnthropicClient(BaseLLMClient):
|
||||
"""Client for Anthropic Claude models."""
|
||||
|
|
@ -16,11 +33,11 @@ class AnthropicClient(BaseLLMClient):
|
|||
"""Return configured ChatAnthropic instance."""
|
||||
llm_kwargs = {"model": self.model}
|
||||
|
||||
for key in ("timeout", "max_retries", "api_key", "max_tokens", "callbacks", "http_client", "http_async_client"):
|
||||
for key in _PASSTHROUGH_KWARGS:
|
||||
if key in self.kwargs:
|
||||
llm_kwargs[key] = self.kwargs[key]
|
||||
|
||||
return ChatAnthropic(**llm_kwargs)
|
||||
return NormalizedChatAnthropic(**llm_kwargs)
|
||||
|
||||
def validate_model(self) -> bool:
|
||||
"""Validate model for Anthropic."""
|
||||
|
|
|
|||
|
|
@ -2,6 +2,25 @@ from abc import ABC, abstractmethod
|
|||
from typing import Any, Optional
|
||||
|
||||
|
||||
def normalize_content(response):
|
||||
"""Normalize LLM response content to a plain string.
|
||||
|
||||
Multiple providers (OpenAI Responses API, Google Gemini 3) return content
|
||||
as a list of typed blocks, e.g. [{'type': 'reasoning', ...}, {'type': 'text', 'text': '...'}].
|
||||
Downstream agents expect response.content to be a string. This extracts
|
||||
and joins the text blocks, discarding reasoning/metadata blocks.
|
||||
"""
|
||||
content = response.content
|
||||
if isinstance(content, list):
|
||||
texts = [
|
||||
item.get("text", "") if isinstance(item, dict) and item.get("type") == "text"
|
||||
else item if isinstance(item, str) else ""
|
||||
for item in content
|
||||
]
|
||||
response.content = "\n".join(t for t in texts if t)
|
||||
return response
|
||||
|
||||
|
||||
class BaseLLMClient(ABC):
|
||||
"""Abstract base class for LLM clients."""
|
||||
|
||||
|
|
|
|||
|
|
@ -2,30 +2,19 @@ from typing import Any, Optional
|
|||
|
||||
from langchain_google_genai import ChatGoogleGenerativeAI
|
||||
|
||||
from .base_client import BaseLLMClient
|
||||
from .base_client import BaseLLMClient, normalize_content
|
||||
from .validators import validate_model
|
||||
|
||||
|
||||
class NormalizedChatGoogleGenerativeAI(ChatGoogleGenerativeAI):
|
||||
"""ChatGoogleGenerativeAI with normalized content output.
|
||||
|
||||
Gemini 3 models return content as list: [{'type': 'text', 'text': '...'}]
|
||||
Gemini 3 models return content as list of typed blocks.
|
||||
This normalizes to string for consistent downstream handling.
|
||||
"""
|
||||
|
||||
def _normalize_content(self, response):
|
||||
content = response.content
|
||||
if isinstance(content, list):
|
||||
texts = [
|
||||
item.get("text", "") if isinstance(item, dict) and item.get("type") == "text"
|
||||
else item if isinstance(item, str) else ""
|
||||
for item in content
|
||||
]
|
||||
response.content = "\n".join(t for t in texts if t)
|
||||
return response
|
||||
|
||||
def invoke(self, input, config=None, **kwargs):
|
||||
return self._normalize_content(super().invoke(input, config, **kwargs))
|
||||
return normalize_content(super().invoke(input, config, **kwargs))
|
||||
|
||||
|
||||
class GoogleClient(BaseLLMClient):
|
||||
|
|
|
|||
|
|
@ -3,20 +3,19 @@ from typing import Any, Optional
|
|||
|
||||
from langchain_openai import ChatOpenAI
|
||||
|
||||
from .base_client import BaseLLMClient
|
||||
from .base_client import BaseLLMClient, normalize_content
|
||||
from .validators import validate_model
|
||||
|
||||
|
||||
class UnifiedChatOpenAI(ChatOpenAI):
|
||||
"""ChatOpenAI subclass that strips temperature/top_p for GPT-5 family models.
|
||||
class NormalizedChatOpenAI(ChatOpenAI):
|
||||
"""ChatOpenAI with normalized content output.
|
||||
|
||||
GPT-5 family models use reasoning natively. temperature/top_p are only
|
||||
accepted when reasoning.effort is 'none'; with any other effort level
|
||||
(or for older GPT-5/GPT-5-mini/GPT-5-nano which always reason) the API
|
||||
rejects these params. Langchain defaults temperature=0.7, so we must
|
||||
strip it to avoid errors.
|
||||
The Responses API returns content as a list of typed blocks
|
||||
(reasoning, text, etc.). This normalizes to string for consistent
|
||||
downstream handling.
|
||||
|
||||
Non-GPT-5 models (GPT-4.1, xAI, Ollama, etc.) are unaffected.
|
||||
Also strips temperature/top_p for GPT-5 family models which use
|
||||
reasoning natively and reject these params.
|
||||
"""
|
||||
|
||||
def __init__(self, **kwargs):
|
||||
|
|
@ -25,9 +24,31 @@ class UnifiedChatOpenAI(ChatOpenAI):
|
|||
kwargs.pop("top_p", None)
|
||||
super().__init__(**kwargs)
|
||||
|
||||
def invoke(self, input, config=None, **kwargs):
|
||||
return normalize_content(super().invoke(input, config, **kwargs))
|
||||
|
||||
# Kwargs forwarded from user config to ChatOpenAI
|
||||
_PASSTHROUGH_KWARGS = (
|
||||
"timeout", "max_retries", "reasoning_effort",
|
||||
"api_key", "callbacks", "http_client", "http_async_client",
|
||||
)
|
||||
|
||||
# Provider base URLs and API key env vars
|
||||
_PROVIDER_CONFIG = {
|
||||
"xai": ("https://api.x.ai/v1", "XAI_API_KEY"),
|
||||
"openrouter": ("https://openrouter.ai/api/v1", "OPENROUTER_API_KEY"),
|
||||
"ollama": (None, None), # base_url comes from config
|
||||
}
|
||||
|
||||
|
||||
class OpenAIClient(BaseLLMClient):
|
||||
"""Client for OpenAI, Ollama, OpenRouter, and xAI providers."""
|
||||
"""Client for OpenAI, Ollama, OpenRouter, and xAI providers.
|
||||
|
||||
For native OpenAI models, uses the Responses API (/v1/responses) which
|
||||
supports reasoning_effort with function tools across all model families
|
||||
(GPT-4.1, GPT-5). Third-party compatible providers (xAI, OpenRouter,
|
||||
Ollama) use standard Chat Completions.
|
||||
"""
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
|
|
@ -43,31 +64,34 @@ class OpenAIClient(BaseLLMClient):
|
|||
"""Return configured ChatOpenAI instance."""
|
||||
llm_kwargs = {"model": self.model}
|
||||
|
||||
if self.provider == "xai":
|
||||
llm_kwargs["base_url"] = "https://api.x.ai/v1"
|
||||
api_key = os.environ.get("XAI_API_KEY")
|
||||
if api_key:
|
||||
llm_kwargs["api_key"] = api_key
|
||||
elif self.provider == "openrouter":
|
||||
llm_kwargs["base_url"] = "https://openrouter.ai/api/v1"
|
||||
api_key = os.environ.get("OPENROUTER_API_KEY")
|
||||
if api_key:
|
||||
llm_kwargs["api_key"] = api_key
|
||||
elif self.provider == "ollama":
|
||||
# Provider-specific base URL and auth
|
||||
if self.provider == "ollama":
|
||||
host = self.base_url or "http://localhost:11434"
|
||||
# Ensure the URL ends with /v1 for OpenAI-compatible endpoint
|
||||
if not host.rstrip("/").endswith("/v1"):
|
||||
host = host.rstrip("/") + "/v1"
|
||||
llm_kwargs["base_url"] = host
|
||||
llm_kwargs["api_key"] = "ollama" # Ollama doesn't require auth
|
||||
llm_kwargs["api_key"] = "ollama"
|
||||
elif self.provider in _PROVIDER_CONFIG:
|
||||
base_url, api_key_env = _PROVIDER_CONFIG[self.provider]
|
||||
llm_kwargs["base_url"] = base_url
|
||||
if api_key_env:
|
||||
api_key = os.environ.get(api_key_env)
|
||||
if api_key:
|
||||
llm_kwargs["api_key"] = api_key
|
||||
elif self.base_url:
|
||||
llm_kwargs["base_url"] = self.base_url
|
||||
|
||||
for key in ("timeout", "max_retries", "reasoning_effort", "api_key", "callbacks", "http_client", "http_async_client"):
|
||||
# Forward user-provided kwargs
|
||||
for key in _PASSTHROUGH_KWARGS:
|
||||
if key in self.kwargs:
|
||||
llm_kwargs[key] = self.kwargs[key]
|
||||
|
||||
return UnifiedChatOpenAI(**llm_kwargs)
|
||||
# Native OpenAI: use Responses API for consistent behavior across
|
||||
# all model families. Third-party providers use Chat Completions.
|
||||
if self.provider == "openai":
|
||||
llm_kwargs["use_responses_api"] = True
|
||||
|
||||
return NormalizedChatOpenAI(**llm_kwargs)
|
||||
|
||||
def validate_model(self) -> bool:
|
||||
"""Validate model for the provider."""
|
||||
|
|
|
|||
Loading…
Reference in New Issue