Simplify Strategist prompt: concrete example instead of rule-chain

The previous prompt enumerated so many constraints that Qwen3-14B
loaded them into its deliberation loop and never actually produced
the final structured line, causing parse failures. This version shows
one concrete filled example (AAPL BUY 75% $289 $283 $298) and tells
the model to substitute; the literal example also primes it to emit
concrete numbers in the same shape instead of the 'ACTION / XX%'
placeholders it was echoing back.

The sanity-fix in the parser still corrects runaway numbers if they
slip through.

crew/tasks.py

@@ -91,34 +91,22 @@ def create_strategy_task(agent: Agent, ticker: str, context: list) -> Task:
     return Task(
         description=(
             f"Synthesize all analysis for {ticker} into a final trading signal. "
-            f"Consider the market conditions, fundamental valuation, technical signals, "
-            f"and risk parameters. Produce a clear BUY, SELL, or HOLD recommendation "
-            f"with a confidence percentage.\n\n"
-            f"CRITICAL PRICE RULES — check each value before you output it:\n"
-            f"  * Use the CURRENT PRICE reported by the Market Scanner's "
-            f"    get_price_change tool as the basis for Entry. Do NOT invent a "
-            f"    price from memory.\n"
-            f"  * For a BUY recommendation: Stop Loss < Entry < Target. "
-            f"    Target should be at least 3% ABOVE Entry; Stop Loss should be "
-            f"    1-3% BELOW Entry. If these inequalities do not hold, re-derive "
-            f"    the numbers before returning.\n"
-            f"  * For a SELL recommendation: Target < Entry < Stop Loss. "
-            f"    Target should be at least 3% BELOW Entry; Stop Loss should be "
-            f"    1-3% ABOVE Entry (to cap loss if the short moves against you).\n"
-            f"  * For a HOLD recommendation: Entry = current price, and Stop Loss / "
-            f"    Target can be left at approximate boundaries of the current range.\n"
-            f"  * The Risk Manager provides ATR-based stop-loss suggestions — "
-            f"    honor them when they satisfy the BUY/SELL inequality above.\n\n"
-            f"Your output MUST follow this exact format:\n"
-            f"{ticker} — ACTION (Confidence: XX%)\n"
-            f"Entry: $XX.XX\n"
-            f"Stop Loss: $XX.XX\n"
-            f"Target: $XX.XX\n"
+            f"Use the current price reported by get_price_change as Entry. "
+            f"For BUY: Stop Loss below Entry, Target above Entry (within 5%). "
+            f"For SELL: Stop Loss above Entry, Target below Entry (within 5%). "
+            f"Keep the response concise — do not deliberate at length.\n\n"
+            f"Output EXACTLY this format on its own lines, with NO extra prose:\n"
+            f"{ticker} — BUY (Confidence: 75%)\n"
+            f"Entry: $289.00\n"
+            f"Stop Loss: $283.00\n"
+            f"Target: $298.00\n"
             f"Reasoning:\n"
-            f"- Market: [summary]\n"
-            f"- Fundamental: [summary]\n"
-            f"- Technical: [summary]\n"
-            f"- Risk: [summary]"
+            f"- Market: [one line]\n"
+            f"- Fundamental: [one line]\n"
+            f"- Technical: [one line]\n"
+            f"- Risk: [one line]\n\n"
+            f"(Substitute the real action, confidence, prices, and reasoning; "
+            f"keep every line on its own line; do not echo these instructions.)"
         ),
         expected_output=(
             f"A trading signal in the format: "