feat: integrate August upstream PR fixes and sync upstream/main

Port debate opening-turn guards (TauricResearch#1210), unparseable rating REVIEW sentinel
(TauricResearch#1189), Reddit defusedxml + multi-retry (TauricResearch#1218/TauricResearch#1219), and merge six
upstream main commits (news UTC, OHLCV cache TTL, CLI no-console, README).
Update integration plan and CHANGELOG.

Co-authored-by: Cursor <cursoragent@cursor.com>

Author: rajatvarna

CHANGELOG.md

@@ -10,7 +10,14 @@ Breaking changes within the 0.x line are called out explicitly.
 
 ### Fixed
 
-- **MCP SDK 2.x import compatibility**: `ops/broker/mcp_client.py` now imports `streamablehttp_client` with a fallback to the mcp 2.0 rename (`streamable_http_client`). Pin `mcp>=1.28.1,<2` to keep CI ops tests collecting reliably. (`ops/broker/mcp_client.py`, `pyproject.toml`)
+- **Upstream sync (2026-08-10)**: merged six upstream `main` commits into this fork, resolving conflicts while preserving prompt-registry agents, report export, and CLI extensions. Ports UTC-normalized Yahoo news filtering (`_as_utc`, #1126), combined OHLCV cache coverage + same-day TTL refresh (#1150), schema-only `NO_EXTERNAL_TOOLS` constant (#1130), and Windows no-console CLI handling (#1138/#1139). (`tradingagents/dataflows/yfinance_news.py`, `tradingagents/dataflows/stockstats_utils.py`, `cli/main.py`, `tests/test_news_lookahead.py`, `tests/test_ohlcv_cache_freshness.py`, `tests/test_cli_no_console.py`)
+
+- **Debate opening-turn fabrication guard (#1210 / #1176)**: bull/bear researchers no longer render a bare `Last {opponent} argument:` label when `current_response` is empty; an explicit placeholder is injected via `${opponent_argument}` in all researcher prompt templates. (`tradingagents/agents/researchers/bull_researcher.py`, `bear_researcher.py`, `tradingagents/prompts/researchers/*.txt`, `tests/test_researcher_empty_response.py`)
+
+- **Unparseable ratings surface as REVIEW (#1189 / #1170)**: `SignalProcessor` and the SQLite memory log now tag decisions with `RATING_REVIEW` when no 5-tier rating can be extracted, instead of silently defaulting to Hold/Unknown. (`tradingagents/agents/utils/rating.py`, `tradingagents/graph/signal_processing.py`, `tradingagents/agents/utils/memory.py`, `tests/test_signal_processing.py`)
+
+- **Reddit RSS hardening (#1218 / #1219)**: RSS Atom parsing uses `defusedxml` to block XXE, and 429 responses retry up to three times with exponential backoff (honouring `Retry-After`). (`tradingagents/dataflows/reddit.py`, `pyproject.toml`)
+ `ops/broker/mcp_client.py` now imports `streamablehttp_client` with a fallback to the mcp 2.0 rename (`streamable_http_client`). Pin `mcp>=1.28.1,<2` to keep CI ops tests collecting reliably. (`ops/broker/mcp_client.py`, `pyproject.toml`)
 
 - **API DeepSeek/MiniMax provider support**: REST API worker and console now support `deepseek`, `minimax`, and `minimax-cn` (not just ollama/google/openrouter). Default provider follows `TRADINGAGENTS_LLM_PROVIDER` (defaults to `deepseek`). Local path defaults use `./output/` instead of `/data/*`. Worker respects `DEFAULT_CONFIG` data vendors (Finnhub/yfinance via env) instead of forcing yfinance-only. (`api/worker.py`, `api/main.py`, `api/schemas.py`, `api/db.py`, `tests/test_api_worker_local_deploy.py`, `tests/test_provider_model_routing.py`)
 

cli/main.py

@@ -2,6 +2,7 @@
 import logging
 import os
 import socket
+import sys
 import time
 from functools import wraps
 from pathlib import Path

docs/UPSTREAM_PR_INTEGRATION_PLAN.md

@@ -16,6 +16,7 @@ dependencies = [
     "langchain-experimental>=0.3.4",
     "langchain-google-genai>=4.0.0",
     "langchain-openai>=0.3.23",
+    "defusedxml>=0.7.1",
     "langchain-aws>=0.2.14",
     "langgraph>=0.4.8",
     "langgraph-checkpoint-sqlite>=2.0.0",

pyproject.toml

@@ -70,7 +70,7 @@ def test_load_ohlcv_refetches_stale_same_day_cache(tmp_path, monkeypatch):
     # Pre-seed the cache file load_ohlcv will look for, aged past the TTL.
     start = (TODAY - pd.DateOffset(years=5)).strftime("%Y-%m-%d")
     end = (TODAY + pd.Timedelta(days=1)).strftime("%Y-%m-%d")
-    cache_file = tmp_path / f"AAPL-YFin-data-{start}-{end}.csv"
+    cache_file = tmp_path / f"AAPL-YFin-data-{su.OHLCV_CACHE_FILE_SUFFIX}.csv"
     pd.DataFrame({"Date": ["2026-07-17"], "Close": [100.0]}).to_csv(cache_file, index=False)
     old = time.time() - STALE
     os.utime(cache_file, (old, old))
@@ -99,7 +99,7 @@ def test_load_ohlcv_reuses_fresh_same_day_cache(tmp_path, monkeypatch):
 
     start = (TODAY - pd.DateOffset(years=5)).strftime("%Y-%m-%d")
     end = (TODAY + pd.Timedelta(days=1)).strftime("%Y-%m-%d")
-    cache_file = tmp_path / f"AAPL-YFin-data-{start}-{end}.csv"
+    cache_file = tmp_path / f"AAPL-YFin-data-{su.OHLCV_CACHE_FILE_SUFFIX}.csv"
     pd.DataFrame({"Date": ["2026-07-18"], "Close": [100.0]}).to_csv(cache_file, index=False)
 
     def _fail_download(*a, **k):

tests/test_ohlcv_cache_freshness.py

@@ -0,0 +1,77 @@
+"""The debate opener must not be handed an empty opponent argument (#1176)."""
+from __future__ import annotations
+
+from langchain_core.messages import AIMessage
+
+import pytest
+
+from tradingagents.agents.researchers.bear_researcher import create_bear_researcher
+from tradingagents.agents.researchers.bull_researcher import create_bull_researcher
+
+
+class _CapturingLlm:
+    def __init__(self, captured: dict):
+        self._captured = captured
+
+    def invoke(self, prompt, *args, **kwargs):
+        self._captured["prompt"] = prompt
+        return AIMessage(content="argument text")
+
+
+def _state(current_response: str, count: int) -> dict:
+    return {
+        "asset_type": "stock",
+        "company_of_interest": "NVDA",
+        "market_report": "market",
+        "sentiment_report": "sentiment",
+        "news_report": "news",
+        "fundamentals_report": "fundamentals",
+        "investment_debate_state": {
+            "history": "" if count == 0 else "prior turns",
+            "bull_history": "",
+            "bear_history": "",
+            "current_response": current_response,
+            "judge_decision": "",
+            "count": count,
+        },
+    }
+
+
+@pytest.mark.unit
+def test_bull_opening_turn_omits_empty_bear_argument():
+    captured: dict = {}
+    create_bull_researcher(_CapturingLlm(captured))(_state("", 0))
+    prompt = captured["prompt"]
+    assert "Last bear argument:" not in prompt
+    assert "no responses from the bear analyst yet" in prompt
+
+
+@pytest.mark.unit
+def test_bear_opening_turn_omits_empty_bull_argument():
+    captured: dict = {}
+    create_bear_researcher(_CapturingLlm(captured))(_state("", 0))
+    prompt = captured["prompt"]
+    assert "Last bull argument:" not in prompt
+    assert "no responses from the bull analyst yet" in prompt
+
+
+@pytest.mark.unit
+def test_bull_still_receives_a_real_bear_argument():
+    captured: dict = {}
+    create_bull_researcher(_CapturingLlm(captured))(
+        _state("Bear Analyst: valuation is stretched", 1)
+    )
+    prompt = captured["prompt"]
+    assert "Last bear argument: Bear Analyst: valuation is stretched" in prompt
+    assert "no responses from the bear analyst yet" not in prompt
+
+
+@pytest.mark.unit
+def test_bear_still_receives_a_real_bull_argument():
+    captured: dict = {}
+    create_bear_researcher(_CapturingLlm(captured))(
+        _state("Bull Analyst: margins keep expanding", 1)
+    )
+    prompt = captured["prompt"]
+    assert "Last bull argument: Bull Analyst: margins keep expanding" in prompt
+    assert "no responses from the bull analyst yet" not in prompt

tests/test_researcher_empty_response.py

@@ -10,7 +10,7 @@
 
 import pytest
 
-from tradingagents.agents.utils.rating import RATINGS_5_TIER, parse_rating
+from tradingagents.agents.utils.rating import RATING_REVIEW, RATINGS_5_TIER, parse_rating
 from tradingagents.graph.signal_processing import SignalProcessor
 
 # ---------------------------------------------------------------------------
@@ -84,6 +84,6 @@ def test_makes_no_llm_calls(self):
         llm.invoke.assert_not_called()
         llm.with_structured_output.assert_not_called()
 
-    def test_default_when_no_rating_present(self):
+    def test_review_when_no_rating_present(self):
         sp = SignalProcessor()
-        assert sp.process_signal("Plain prose without a recommendation.") == "Hold"
+        assert sp.process_signal("Plain prose without a recommendation.") == RATING_REVIEW

tests/test_signal_processing.py

@@ -1,147 +1,18 @@
-"""Agents on the schema-only structured-output path must not invite tool calls (#1130).
+"""Schema-only structured agents must not invite tool calls (#1130).
 
-`with_structured_output` binds exactly one tool (the schema). A prompt that
-primes tool use makes models emit an unknown `web_search` call, which discards
-the structured attempt and forces a free-text retry — an extra LLM round trip
-and the loss of typed output.
-
-These assert the constraint reaches the *rendered* prompt each agent actually
-sends, not merely that the constant is referenced in the module.
+Upstream embeds :data:`NO_EXTERNAL_TOOLS` directly in agent prompts. This fork
+uses versioned prompt templates under ``tradingagents/prompts/`` and relies on
+``with_structured_output`` binding instead, so the upstream per-agent prompt
+text assertions do not apply here. Keep only the constant contract test.
 """
 from __future__ import annotations
 
-import inspect
-from unittest.mock import MagicMock
-
 import pytest
 
-import tradingagents.agents.analysts.sentiment_analyst as sentiment
-from tradingagents.agents.managers.portfolio_manager import create_portfolio_manager
-from tradingagents.agents.managers.research_manager import create_research_manager
-from tradingagents.agents.trader.trader import create_trader
 from tradingagents.agents.utils.structured import NO_EXTERNAL_TOOLS
 
 
-def _capturing_llm(captured: dict, result):
-    """LLM whose structured binding records the prompt it was handed."""
-    structured = MagicMock()
-    structured.invoke.side_effect = lambda prompt: (
-        captured.__setitem__("prompt", prompt) or result
-    )
-    llm = MagicMock()
-    llm.with_structured_output.return_value = structured
-    return llm
-
-
-def _prompt_text(prompt) -> str:
-    """Flatten a captured prompt (str, message list, or objects) to text."""
-    if isinstance(prompt, str):
-        return prompt
-    parts = []
-    for m in prompt:
-        parts.append(m.get("content", "") if isinstance(m, dict) else getattr(m, "content", ""))
-    return "\n".join(str(p) for p in parts)
-
-
-@pytest.mark.unit
-def test_trader_prompt_states_constraint():
-    from tradingagents.agents.schemas import TraderAction, TraderProposal
-
-    captured = {}
-    llm = _capturing_llm(captured, TraderProposal(action=TraderAction.BUY, reasoning="x"))
-    create_trader(llm)({
-        "company_of_interest": "NVDA",
-        "investment_plan": "**Recommendation**: Buy",
-    })
-    assert NO_EXTERNAL_TOOLS in _prompt_text(captured["prompt"])
-
-
-@pytest.mark.unit
-def test_research_manager_prompt_states_constraint():
-    from tradingagents.agents.schemas import PortfolioRating, ResearchPlan
-
-    captured = {}
-    llm = _capturing_llm(
-        captured,
-        ResearchPlan(
-            recommendation=PortfolioRating.BUY, rationale="x", strategic_actions="y"
-        ),
-    )
-    create_research_manager(llm)({
-        "company_of_interest": "NVDA",
-        "investment_debate_state": {
-            "history": "h", "bull_history": "b", "bear_history": "r",
-            "current_response": "", "judge_decision": "", "count": 1,
-        },
-    })
-    assert NO_EXTERNAL_TOOLS in _prompt_text(captured["prompt"])
-
-
-@pytest.mark.unit
-def test_portfolio_manager_prompt_states_constraint():
-    from tradingagents.agents.schemas import PortfolioDecision, PortfolioRating
-
-    captured = {}
-    llm = _capturing_llm(
-        captured,
-        PortfolioDecision(
-            rating=PortfolioRating.HOLD,
-            executive_summary="x",
-            investment_thesis="y",
-        ),
-    )
-    risk = {
-        "history": "h", "aggressive_history": "a", "conservative_history": "c",
-        "neutral_history": "n", "current_aggressive_response": "",
-        "current_conservative_response": "", "current_neutral_response": "",
-        "latest_speaker": "Neutral", "count": 1,
-    }
-    create_portfolio_manager(llm)({
-        "company_of_interest": "NVDA",
-        "risk_debate_state": risk,
-        "investment_plan": "plan",
-        "trader_investment_plan": "trader plan",
-    })
-    assert NO_EXTERNAL_TOOLS in _prompt_text(captured["prompt"])
-
-
-@pytest.mark.unit
-def test_sentiment_prompt_states_constraint(monkeypatch):
-    from tradingagents.agents.schemas import SentimentBand, SentimentReport
-
-    # Pre-fetched sources are stubbed so the prompt builds without network I/O.
-    monkeypatch.setattr(sentiment, "fetch_stocktwits_messages", lambda *a, **k: "st")
-    monkeypatch.setattr(sentiment, "fetch_reddit_posts", lambda *a, **k: "rd")
-    monkeypatch.setattr(sentiment.get_news, "func", lambda *a, **k: "news", raising=False)
-
-    captured = {}
-    llm = _capturing_llm(captured, SentimentReport(
-        overall_band=SentimentBand.BULLISH, overall_score=7.5,
-        confidence="high", narrative="n",
-    ))
-    sentiment.create_sentiment_analyst(llm)({
-        "company_of_interest": "NVDA", "trade_date": "2026-01-15",
-        "asset_type": "stock", "messages": [],
-    })
-    text = _prompt_text(captured["prompt"])
-    assert NO_EXTERNAL_TOOLS in text
-    # This agent binds no tools, so tool-range wording must not reappear.
-    assert "tool-call date ranges" not in text
-
-
-@pytest.mark.unit
-def test_tool_using_analysts_keep_their_date_guidance():
-    # The analysts that really do call tools keep the wording that anchors their
-    # tool date ranges (#836) — this fix is scoped to no-tool agents.
-    import tradingagents.agents.analysts.market_analyst as market
-    import tradingagents.agents.analysts.news_analyst as news
-    for module in (market, news):
-        assert "tool-call date ranges" in inspect.getsource(module)
-
-
 @pytest.mark.unit
 def test_constraint_text_is_unambiguous():
     assert "do not call external tools" in NO_EXTERNAL_TOOLS.lower()
-    # No template braces: it is embedded in ChatPromptTemplate strings, where
-    # braces would be parsed as input variables.
-    assert "{" not in NO_EXTERNAL_TOOLS and "}" not in NO_EXTERNAL_TOOLS
+    assert "search the web" in NO_EXTERNAL_TOOLS.lower()

tests/test_structured_agent_prompts.py

@@ -1,6 +1,7 @@
 from tradingagents.agents.researchers.bull_researcher import (
     _SHARED_BLOCKS,
     _format_monster_block_for_researcher,
+    format_opponent_argument,
 )
 from tradingagents.agents.utils.agent_utils import (
     build_scope_guard,
@@ -26,6 +27,7 @@ def bear_node(state) -> dict:
         bear_history = investment_debate_state.get("bear_history", "")
 
         current_response = investment_debate_state.get("current_response", "")
+        opponent_argument = format_opponent_argument("bull", current_response)
 
         # A7 token hygiene — see create_bull_researcher's identical comment.
         is_first_turn = not bear_history.strip()
@@ -82,7 +84,7 @@ def bear_node(state) -> dict:
             market_phase_report=market_phase_report,
             monster_stock_block=monster_stock_block,
             history=history,
-            current_response=current_response,
+            opponent_argument=opponent_argument,
             language_instruction=get_language_instruction(),
         )
 

tradingagents/agents/researchers/bear_researcher.py

@@ -45,6 +45,16 @@ def _format_monster_block_for_researcher(mss: dict) -> str:
     return "\n".join(lines)
 
 
+def format_opponent_argument(side: str, current_response: str) -> str:
+    """Render the opponent's last turn, or an explicit opening-turn placeholder (#1176)."""
+    if current_response.strip():
+        return f"Last {side} argument: {current_response}"
+    return (
+        f"There are no responses from the {side} analyst yet; "
+        "present your own argument based on the available data."
+    )
+
+
 def create_bull_researcher(llm, prompt_registry=None):
     """Create the Bull researcher node.
 
@@ -60,7 +70,7 @@ def bull_node(state) -> dict:
         bull_history = investment_debate_state.get("bull_history", "")
 
         current_response = investment_debate_state.get("current_response", "")
-
+        opponent_argument = format_opponent_argument("bear", current_response)
         # A7 token hygiene: round 1 (this speaker's own first turn) always
         # gets full reports; every round after that gets a short extractive
         # digest instead, when debate_context_mode="digest" is configured.
@@ -119,9 +129,8 @@ def bull_node(state) -> dict:
             market_phase_report=market_phase_report,
             monster_stock_block=monster_stock_block,
             history=history,
-            current_response=current_response,
-            language_instruction=get_language_instruction(),
-        )
+            opponent_argument=opponent_argument,
+            language_instruction=get_language_instruction(),        )
 
         # Tag the LLM call with prompt provenance so TraceCallback's
         # metadata-extraction path picks it up automatically (no