asgardeo
MCP server to interact with your Asgardeo organization through LLM tools
claude mcp add --transport stdio asgardeo-asgardeo-mcp-server <absolute path to the asgardeo-mcp executable, e.g., /Users/<user directory>/<repository path>/asgardeo-mcp-server/asgardeo-mcp> \ --env BASE_URL="https://api.asgardeo.io/t/<asgardeo organization>" \ --env CLIENT_ID="<client ID>" \ --env CLIENT_SECRET="<client secret>"
How to use
The Asgardeo MCP Server provides a set of tools to manage your Asgardeo organization or WSO2 Identity Server deployment through natural language prompts and structured tool calls. The server exposes application management capabilities such as listing, creating, updating, and retrieving details for applications; API resource management; and login flow configuration driven by prompts. These tools are designed to be invoked by MCP clients (e.g., Claude Desktop, VS Code MCP integration, Cursor) to perform configuration tasks across your identity platform via the Asgardeo APIs. The included commands like list_applications, create_single_page_app, create_m2m_app, get_application_by_name, and update_login_flow enable end-to-end lifecycle management of your applications and their authentication configurations. The server supports both standard application management APIs and related resources, making it possible to script or chat your desired configuration changes.
To use the MCP server, you typically configure an MCP client with the server executable path and environment credentials (BASE_URL, CLIENT_ID, CLIENT_SECRET). Then you can issue tool calls or natural language prompts that map to the available tools, and the MCP server will translate those prompts into API requests to your Asgardeo or WSO2 Identity Server deployment. This enables you to perform tasks such as listing all applications, creating new applications (including M2M and various web/mobile variants), fetching application details by name or client ID, and updating login flows or OAuth configurations through conversational interactions.
How to install
Prerequisites:
- Go installed (for building the MCP server from source)
- Git installed
- Access to the Asgardeo organization and credentials (for BASE_URL, CLIENT_ID, CLIENT_SECRET)
Installation steps:
- Clone the repository:
git clone https://github.com/asgardeo/asgardeo-mcp-server.git
- Change to the repository directory:
cd asgardeo-mcp-server
- Install dependencies and tidy modules (Go):
go mod tidy
- Build the executable:
go build -o asgardeo-mcp
- Verify the binary exists:
ls -l asgardeo-mcp
- Configure MCP client integrations by providing the required environment variables (BASE_URL, CLIENT_ID, CLIENT_SECRET) as described in the mcp_config example. You can run the server directly or integrate it into your MCP client tooling as demonstrated in the documentation.
Additional notes
Tips and considerations:
- If you are using a WSO2 Identity Server backend, set the PRODUCT_MODE environment variable to wso2is and adjust BASE_URL to your WSO2 IS base URL where tenants are hosted (e.g., https://<host>/t/<tenant>). You may also need to configure CA certificates for SSL in environments with internal or self-signed certificates by setting CERT_PATH to the CA certificate file.
- The MCP server is labeled experimental; avoid using it in production environments without thorough validation.
- Ensure your MCP client is configured with the correct command path to the asgardeo-mcp executable and that the environment variables BASE_URL, CLIENT_ID, and CLIENT_SECRET are kept secure.
- For local development or internal networks, consider using proper TLS certificates and validating certificates to prevent SSL errors during API calls.
Related MCP Servers
mcp-agent
Build effective agents using Model Context Protocol and simple workflow patterns
adk-go
An open-source, code-first Go toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
kagent
Cloud Native Agentic AI | Discord: https://bit.ly/kagentdiscord
sdk-typescript
A model-driven approach to building AI agents in just a few lines of code.
station
Station is our open-source runtime that lets teams deploy agents on their own infrastructure with full control.
mesh
One secure endpoint for every MCP server. Deploy anywhere.