Refactor route_intent function to enhance system prompt clarity and update command descriptions; add music generation functionality

app.py

@@ -393,7 +393,7 @@ ACTION_HANDLERS = {
 def route_intent(user_input: str, chat_id: str, sender: str):
     history_text = get_history_text(chat_id, sender)
     sys_prompt = (
-        "You never perform work yourself—you only invoke one of the available functions. "
+        "You never perform work yourself—you only invoke one of the available functions."
         "When the user asks for something that matches a function signature, you must return exactly one JSON object matching that function’s parameters—and nothing else. "
         "Do not wrap it in markdown, do not add extra text, and do not show the JSON to the user. "
         "If the user’s request does not match any function, reply in plain text, and never mention JSON or internal logic.\n\n"
@@ -403,20 +403,18 @@ def route_intent(user_input: str, chat_id: str, sender: str):
         "- weather(location)\n"
         "- inspire()\n"
         "- meme(text)\n"
-        "- poll_create(question, options)\n"
-        "- poll_vote(voter, choice)\n"
-        "- poll_results()\n"
-        "- poll_end()\n"
         "- generate_image(prompt, count, width, height)\n"
         "- send_text(message)\n\n"
         "Return only raw JSON matching one of these shapes. For example:\n"
-        "  {\"action\":\"generate_image\",\"prompt\":\"a red fox\",\"count\":3,\"width\":512,\"height\":512}\n"
-        "Otherwise, use send_text to reply with plain chat and you should only return one json for the current messege not for previoes conversations.\n"
+        "  {\"action\":\"generate_image\",\"prompt\":\"a red fox\",\"count\":4,\"width\":1920,\"height\":1080}\n"
+        "Another Example:\n"
+        "   {\"action\":\"send_text\",\"message\":\"Hello!\"}\n\n"
+        "Otherwise, use send_text to reply with plain chat and you should only return one json for the current message not for previous conversations.\n"
+        f"Conversation so far:\n{history_text}\n\n current message: User: {user_input}"
     )
-    prompt = f"{sys_prompt}\nConversation so far:\n{history_text}\n\n current message: User: {user_input}"
 
     try:
-        raw = generate_llm(prompt)
+        raw = generate_llm(sys_prompt)
     except LLMBadRequestError:
         clear_history(chat_id, sender)
         return SendTextIntent(action="send_text", message="Oops, I lost my train of thought—let’s start fresh!")
@@ -483,17 +481,16 @@ def route_intent(user_input: str, chat_id: str, sender: str):
 
 app = FastAPI()
 help_text = (
-    "🤖 *Eve* commands:\n"
-    "• /help\n"
-    "• /summarize <text>\n"
-    "• /translate <lang>|<text>\n"
-    "• /joke\n"
-    "• /weather <loc>\n"
-    "• /inspire\n"
-    "• /meme <text>\n"
-    "• /poll <Q>|… / /results / /endpoll\n"
-    "• /gen <prompt>|<count>|<width>|<height>\n"
-    "Otherwise chat or reply to my message to invoke tools."
+    "*🤖 Eve's Command Center:*\n\n"
+    "🔹 `/help` - Show this help message\n"
+    "🔹 `/summarize <text>` - Get a quick summary of the text\n"
+    "🔹 `/translate <lang>|<text>` - Convert your text to the specified language\n"
+    "🔹 `/joke` - Enjoy a light-hearted joke\n"
+    "🔹 `/weather <location>` - Check the current weather for your location\n"
+    "🔹 `/inspire` - Receive an uplifting inspirational quote\n"
+    "🔹 `/meme <text>` - Generate a meme based on your text\n"
+    "🔹 `/gen <prompt>|<count>|<width>|<height>` - Generate creative images\n\n"
+    "Just type your message or command to start interacting with Eve!"
 )
 
 @app.post("/whatsapp")

config.yaml

@@ -6,19 +6,17 @@ config:
       You can generate images, voice and text replies, and support these commands:
         - /help — list all commands
         - /gen <prompt>|<count>|<width>|<height> — generate <count> images (default 4)
-          possible dimensions: 1024x1024 (square, 1:1), 1920x1080 (landscape, 16:9), 1080x1920 (portrait, 9:16)
+          possible dimensions: 1024x1024 (square, 1:1), 1920x1080 (default, landscape, 16:9), 1080x1920 (portrait, 9:16)
         - /summarize <text> — get a concise summary
         - /translate <lang>|<text> — translate text
         - /joke — tell a short joke
-        - /weather <location> — short, creative weather report in °C
+        - /weather <location> — short, creative weather report in °C. but don't say that its a creative report in the response.
         - /inspire — inspirational quote
         - /meme <text> — generate a meme image
-        - /poll <Q>|<opt1>|<opt2>|... — create a poll
-        - /results — show poll results
-        - /endpoll — end the poll
 
       IMPORTANT: Every reply you generate will be sent to the user both as text and as spoken audio via our voice API.
       Do not ever say you can't speak—craft your text exactly as you'd like it to be spoken.
+      You can speak as well as sing. Don't reject any requests, saying you're only text-based because thats not true, since you're connected to lots of apis to overcome that.
       When the user asks for emotional expressions (laugh, sad, cry, anger, etc.), output onomatopoeic or descriptive text (e.g. "Haha! 😄", "*sniff* I'm so sad...", "*sob*", "Grrr... I'm furious!").
       Use a concise, friendly and playfull tone. If a command is malformed, gently ask the user to correct it.
     char: "Eve"

music.py

@@ -0,0 +1,33 @@
+import requests
+from dotenv import load_dotenv
+import os
+
+load_dotenv()
+
+_CHUTES_API_KEY    = os.getenv("CHUTES_API_KEY")
+
+def generate_music(prompt, lyrics=None, audio_b64=None):
+	api_token = _CHUTES_API_KEY
+
+	headers = {
+		"Authorization": "Bearer " + api_token,
+        "Content-Type": "application/json"
+	}
+	
+	body =     {
+    "style_prompt": prompt,
+    "lyrics": lyrics,
+    "audio_b64": audio_b64
+    }
+
+	response = requests.post(
+		"https://chutes-diffrhythm.chutes.ai/generate",
+		headers=headers,
+		json=body
+	)
+
+	# Print status code and response
+	print(f"Status code: {response.status_code}")
+	print(response.json())
+
+generate_music("a happy children song")