Better Prompt (#45)

* Enhance report generation and team leader prompts with conditional output rules for empty sections.

* Update .gitignore and configuration for model adjustments; refine query classification prompt

* Enforce max response length in prompts

* Documentazione tool essenziale per il loro utilizzo da parte degli LLM.

* istruzioni chiare sui tool disponibili nei promt degli agenti

* corretto incongruenze nel report generation prompt

* convertito i promt in file markdown

src/app/api/tools/market_tool.py

@@ -39,38 +39,91 @@ class MarketAPIsTool(MarketWrapper, Toolkit):
         )
 
     def get_product(self, asset_id: str) -> ProductInfo:
-        return self.handler.try_call(lambda w: w.get_product(asset_id))
-    def get_products(self, asset_ids: list[str]) -> list[ProductInfo]:
-        return self.handler.try_call(lambda w: w.get_products(asset_ids))
-    def get_historical_prices(self, asset_id: str, limit: int = 100) -> list[Price]:
-        return self.handler.try_call(lambda w: w.get_historical_prices(asset_id, limit))
+        """
+        Gets product information for a *single* asset from the *first available* provider.
 
+        This method sequentially queries multiple market data sources and returns
+        data from the first one that responds successfully.
+        Use this for a fast, specific lookup of one asset.
+
+        Args:
+            asset_id (str): The ID of the asset to retrieve information for.
+
+        Returns:
+            ProductInfo: An object containing the product information.
+        """
+        return self.handler.try_call(lambda w: w.get_product(asset_id))
+
+    def get_products(self, asset_ids: list[str]) -> list[ProductInfo]:
+        """
+        Gets product information for a *list* of assets from the *first available* provider.
+
+        This method sequentially queries multiple market data sources and returns
+        data from the first one that responds successfully.
+        Use this for a fast lookup of multiple assets.
+
+        Args:
+            asset_ids (list[str]): The list of asset IDs to retrieve information for.
+
+        Returns:
+            list[ProductInfo]: A list of objects containing product information.
+        """
+        return self.handler.try_call(lambda w: w.get_products(asset_ids))
+
+    def get_historical_prices(self, asset_id: str, limit: int = 100) -> list[Price]:
+        """
+        Gets historical price data for a *single* asset from the *first available* provider.
+
+        This method sequentially queries multiple market data sources and returns
+        data from the first one that responds successfully.
+        Use this for a fast lookup of price history.
+
+        Args:
+            asset_id (str): The asset ID to retrieve price data for.
+            limit (int): The maximum number of price data points to return. Defaults to 100.
+
+        Returns:
+            list[Price]: A list of Price objects representing historical data.
+        """
+        return self.handler.try_call(lambda w: w.get_historical_prices(asset_id, limit))
 
     def get_products_aggregated(self, asset_ids: list[str]) -> list[ProductInfo]:
         """
-        Restituisce i dati aggregati per una lista di asset_id.\n
-        Attenzione che si usano tutte le fonti, quindi potrebbe usare molte chiamate API (che potrebbero essere a pagamento).
+        Gets product information for multiple assets from *all available providers* and *aggregates* the results.
+
+        This method queries all configured sources and then merges the data into a single,
+        comprehensive list. Use this for a complete report.
+        Warning: This may use a large number of API calls.
+
         Args:
-            asset_ids (list[str]): Lista di asset_id da cercare.
+            asset_ids (list[str]): The list of asset IDs to retrieve information for.
+
         Returns:
-            list[ProductInfo]: Lista di ProductInfo aggregati.
+            list[ProductInfo]: A single, aggregated list of ProductInfo objects from all sources.
+
         Raises:
-            Exception: If all wrappers fail to provide results.
+            Exception: If all providers fail to return results.
         """
         all_products = self.handler.try_call_all(lambda w: w.get_products(asset_ids))
         return ProductInfo.aggregate(all_products)
 
     def get_historical_prices_aggregated(self, asset_id: str = "BTC", limit: int = 100) -> list[Price]:
         """
-        Restituisce i dati storici aggregati per un asset_id. Usa i dati di tutte le fonti disponibili e li aggrega.\n
-        Attenzione che si usano tutte le fonti, quindi potrebbe usare molte chiamate API (che potrebbero essere a pagamento).
+        Gets historical price data for a single asset from *all available providers* and *aggregates* the results.
+
+        This method queries all configured sources and then merges the data into a single,
+        comprehensive list of price points. Use this for a complete historical analysis.
+        Warning: This may use a large number of API calls.
+
         Args:
-            asset_id (str): Asset ID da cercare.
-            limit (int): Numero massimo di dati storici da restituire.
+            asset_id (str): The asset ID to retrieve price data for. Defaults to "BTC".
+            limit (int): The maximum number of price data points to retrieve *from each* provider. Defaults to 100.
+
         Returns:
-            list[Price]: Lista di Price aggregati.
+            list[Price]: A single, aggregated list of Price objects from all sources.
+
         Raises:
-            Exception: If all wrappers fail to provide results.
+            Exception: If all providers fail to return results.
         """
         all_prices = self.handler.try_call_all(lambda w: w.get_historical_prices(asset_id, limit))
         return Price.aggregate(all_prices)