oic_mcp is a Python FastAPI-based server that acts as a Model Context Protocol (MCP) endpoint, providing comprehensive monitoring and inspection capabilities for Oracle Integration Cloud (OIC) environments. It enables users to programmatically access and analyze OIC integrations, packages, connections, agents, runtime metrics, and crucial design-time information. The server leverages OAuth2 client credentials for secure authentication with IDCS/IAM and interacts with OIC REST APIs using `httpx`. Designed with Large Language Models (LLMs) in mind, all outputs are JSON-first, ensuring structured and easily consumable data, and it even supports offline analysis through local design JSON overrides.
Key Features
01OAuth2 Client Credentials with automatic token refresh for secure access
02Support for offline design-time analysis via local JSON files
03Over 20 read-only tools for comprehensive OIC data inspection
04JSON-first outputs with summaries optimized for LLM consumption
05FastAPI WebSocket server implementing MCP for OIC interactions
060 GitHub stars
Use Cases
01Providing structured OIC metadata and runtime details for consumption by Large Language Models (LLMs) and AI applications.
02Performing offline design-time analysis of OIC integration flows, steps, and mappings.
03Monitoring and inspecting Oracle Integration Cloud (OIC) environments programmatically.
