Add ui agent (#343)

* add uipath agent

* readme update

mm_agents/uipath/llm_client.py

@@ -0,0 +1,44 @@
+import os
+import requests
+
+def send_messages(payload):
+    # URL to your proxy for calling LLMs
+    proxy_url = ""
+    api_key = os.getenv("SERVICE_KEY")
+    
+    # Can be directly replaced with code for calling Azure endpoint as in:
+    #.env config example :
+    # AZURE_OPENAI_API_BASE=YOUR_API_BASE
+    # AZURE_OPENAI_DEPLOYMENT=YOUR_DEPLOYMENT
+    # AZURE_OPENAI_API_VERSION=YOUR_API_VERSION
+    # AZURE_OPENAI_MODEL=gpt-4o-mini
+    # AZURE_OPENAI_API_KEY={{YOUR_API_KEY}}
+    # AZURE_OPENAI_ENDPOINT=${AZURE_OPENAI_API_BASE}/openai/deployments/${AZURE_OPENAI_DEPLOYMENT}/chat/completions?api-version=${AZURE_OPENAI_API_VERSION}
+
+
+    # Load environment variables
+    # load_dotenv()
+    # api_key = os.getenv('AZURE_OPENAI_API_KEY')
+    # openai_endpoint = os.getenv('AZURE_OPENAI_ENDPOINT')
+    # #logger.info("Openai endpoint: %s", openai_endpoint)
+
+    # headers = {
+    #     "Content-Type": "application/json",
+    #     "api-key": api_key
+    # }
+    # response = requests.post(
+    #     openai_endpoint,
+    #     headers=headers,
+    #     json=payload
+    # )
+            
+    headers = {
+        "Content-Type": "application/json",
+        "X-API-KEY": api_key
+    }
+    retries = 3
+    for attempt in range(retries):
+        response = requests.post(proxy_url, headers=headers, json=payload)
+        if response.status_code == 200:
+            return response.json()["choices"][0]["message"]["content"]
+    return None