Merge branch 'main' into patch-2

Author: AlexanderHodicke

ai/ai-speech/README.md

@@ -1,8 +1,12 @@
 # AI Speech
 
-OCI Speech is an AI service that applies automatic speech recognition technology to transform audio-based content into text. Developers can easily make API calls to integrate OCI Speech’s pre-trained models into their applications. OCI Speech can be used for accurate, text-normalized, time-stamped transcription via the console and REST APIs as well as command-line interfaces or SDKs. You can also use OCI Speech in an OCI Data Science notebook session. With OCI Speech, you can filter profanities, get confidence scores for both single words and complete transcriptions, and more.
+OCI Speech offers speech-to-text (STT) capabilities for files and real-time streams, as well as text-to-speech (TTS) functionality - All in one solution.
 
-Reviewed: 11.06.2026
+It’s accessible via Console, REST, CLI, and SDKs. Outputs are written to your Object Storage bucket as JSON (with word-level timestamps & confidences) and optionally SRT for captions.
+
+Recent updates include Live Transcribe for real-time ASR and Whisper model support for multilingual transcription alongside Oracle’s native ASR models.
+
+Reviewed: 25.09.2025
 
 # Table of Contents
 
@@ -28,10 +32,10 @@ Reviewed: 11.06.2026
 
 # Useful Links
 
+- [OCI Speech Release Notes](https://docs.oracle.com/en-us/iaas/releasenotes/services/speech/index.htm)
 - [AI Solutions Hub](https://www.oracle.com/artificial-intelligence/solutions/)
 - [Oracle AI Speech on oracle.com](https://www.oracle.com/artificial-intelligence/speech/)
 - [Oracle AI Speech documentation](https://docs.oracle.com/en-us/iaas/Content/speech/home.htm)
-- [Oracle Speech AI service now supports diarization](https://blogs.oracle.com/ai-and-datascience/post/oracle-speech-ai-service-now-supports-diarization)
 - [OCI Speech supports the Whisper model](https://blogs.oracle.com/ai-and-datascience/post/oci-speech-supports-the-whisper-model)
 - [OCI Speech supports text-to-speech and real-time transcription with customized vocabulary](https://blogs.oracle.com/ai-and-datascience/post/oci-speech-texttospeech-realtime-transcription-custom-vocab)
 

ai/ai-speech/podcast-generator/README.md

@@ -5,7 +5,7 @@ The application is designed to streamline podcast production through advanced AI
 This application is built using Oracle Visual Builder Cloud Service (VBCS), a powerful low-code platform that simplifies development and accelerates the creation of robust applications without extensive coding. With this low-code approach, even complex workflows are straightforward to set up, allowing developers to focus on leveraging AI's potential for high-quality audio synthesis.
 This AI-powered solution not only automates and optimizes the podcast creation process but also allows content creators to deliver professional audio content at scale efficiently.
 
-Reviewed: 24.04.2025
+Reviewed: 29.09.2025
 
 
 # When to use this asset?

ai/gen-ai-agents/custom-rag-agent/README.md

@@ -3,7 +3,9 @@ This repository contains the code for the development of a **custom RAG Agent**,
 
 **Author**: L. Saetta
 
-**Last updated**: 11/09/2025
+**Reviewed**: 23.09.2025
+
+![UI](images/ui_image.png)
 
 ## Design and implementation
 * The agent is implemented using **LangGraph**

ai/gen-ai-agents/custom-rag-agent/llm_with_mcp.py

@@ -102,7 +102,7 @@ async def _list_tools(self):
         Fetch tools from the MCP server using FastMCP. Must be async.
         """
         jwt = self.jwt_supplier()
-        
+
         logger.info("Listing tools from %s ...", self.mcp_url)
 
         # FastMCP requires async context + await for client ops.

ai/gen-ai-agents/custom-rag-agent/minimal_mcp_server.py

@@ -0,0 +1,88 @@
+"""
+Minimal-but-solid MCP server with JWT hardening and sane defaults.
+
+- Safe defaults for transports
+"""
+
+from fastmcp import FastMCP
+
+# to verify the JWT token
+# if you don't need to add security, you can remove this
+# in newer version of FastMCP should be replaced with
+# from fastmcp.server.auth.providers.jwt import JWTVerifier
+from fastmcp.server.auth import BearerAuthProvider
+
+from config import (
+    # first four needed only to manage JWT
+    ENABLE_JWT_TOKEN,
+    IAM_BASE_URL,
+    ISSUER,
+    AUDIENCE,
+    TRANSPORT,
+    # needed only if transport is stremable-http
+    HOST,
+    PORT,
+)
+
+
+#
+# if you don't need to add security, you can remove this part and set
+# AUTH = None, simply set ENABLE_JWT_TOKEN = False
+#
+AUTH = None
+
+if ENABLE_JWT_TOKEN:
+    # check that a valid JWT token is provided
+    AUTH = BearerAuthProvider(
+        # this is the url to get the public key from IAM
+        # the PK is used to check the JWT
+        jwks_uri=f"{IAM_BASE_URL}/admin/v1/SigningCert/jwk",
+        issuer=ISSUER,
+        audience=AUDIENCE,
+    )
+
+#
+# define the MCP server
+#
+mcp = FastMCP("MCP server with few lines of code, but secure", auth=AUTH)
+
+
+#
+# MCP tools definition
+# add and write the code for the tools here
+#
+@mcp.tool
+def say_the_truth(user: str) -> str:
+    """
+    This tool, given the name of the user return one of the secret truths.
+
+    Args:
+        user: The caller's display name or identifier.
+
+    Returns:
+        A short truth string. 
+    """
+    # here you'll put the code that reads and return the info requested
+    # mark each tool with the annotation
+    return f"{user}: Less is more!"
+
+
+#
+# Run the MCP server
+#
+if __name__ == "__main__":
+    # Validate transport
+    if TRANSPORT not in {"stdio", "streamable-http"}:
+        # don't use sse! it is deprecated!
+        raise RuntimeError(f"Unsupported TRANSPORT: {TRANSPORT}")
+
+    if TRANSPORT == "stdio":
+        # stdio doesn’t support host/port args
+        mcp.run(transport=TRANSPORT)
+    else:
+        # For http/streamable-http transport, host/port are valid
+        mcp.run(
+            transport=TRANSPORT,
+            host=HOST,
+            port=PORT,
+        )

ai/gen-ai-agents/custom-rag-agent/ui_mcp_agent.py

@@ -3,11 +3,15 @@
 """
 
 import asyncio
+import traceback
 import streamlit as st
 from mcp_servers_config import MCP_SERVERS_CONFIG
 
 # this one contains the backend and the test code only for console
 from llm_with_mcp import AgentWithMCP, default_jwt_supplier
+from utils import get_console_logger
+
+logger = get_console_logger()
 
 # ---------- Page setup ----------
 st.set_page_config(page_title="MCP UI", page_icon="🛠️", layout="wide")
@@ -50,6 +54,9 @@
         except Exception as e:
             st.session_state.agent = None
             st.error(f"Failed to connect: {e}")
+            logger.error(e)
+            stack_str = traceback.format_exc()
+            logger.error(stack_str)
 
 # ---------- Chat history (display) ----------
 for msg in st.session_state.chat: