https://oracle.github.io/ai-optimizer/main/advanced/source_code/index.htmlThe AI Optimizer can export your tested configuration as a downloadable source code template, ready to run as a standalone application. These templates expose the RAG chatbot as an OpenAI API-compatible REST endpoint using either LangChain MCP or Spring AI. Templates are available from the Settings page when your configuration uses either Ollama or OpenAI as providers for both chat and embedding models.Hugoen-ushttps://oracle.github.io/ai-optimizer/main/advanced/source_code/langchain_mcp/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://oracle.github.io/ai-optimizer/main/advanced/source_code/langchain_mcp/index.htmlIntroduction to the MCP Server for a tested AI Optimizer & Toolkit configuration This document describes how to re-use the configuration tested in the AI Optimizer and expose it as an MCP tool. The MCP tool can be consumed locally by Claude Desktop or deployed as a remote MCP server using the Python/LangChain framework. This early-stage implementation supports communication via the stdio and sse transports, enabling interaction between the agent dashboard (for example, Claude Desktop) and the exported RAG tool.https://oracle.github.io/ai-optimizer/main/advanced/source_code/springai/index.htmlMon, 01 Jan 0001 00:00:00 +0000https://oracle.github.io/ai-optimizer/main/advanced/source_code/springai/index.htmlAfter having downloaded and unzipped the SpringAI file from the Settings screen, you can open and set the latest two things in the code to be executed. For the detailed description, please refer to the README.md file included Prerequisites Before using a microservice that exploit OpenAI API, make sure you have a developer token from OpenAI. To do this, create an account at OpenAI and generate the token at API Keys.