Get the FREE Ultimate OpenClaw Setup Guide →

casibase

⚡️AI Cloud OS: Open-source enterprise-level AI knowledge base and MCP (model-context-protocol)/A2A (agent-to-agent) management platform with admin UI, user management and Single-Sign-On⚡️, supports ChatGPT, Claude, Llama, Ollama, HuggingFace, etc., chat bot demo: https://ai.casibase.com, admin UI demo: https://ai-admin.casibase.com

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio casibase-casibase docker run -d -p 8000:80 --name casibase casbin/casibase \
  --env CASIBASE_ADMIN_EMAIL="admin@example.com" \
  --env CASIBASE_DATABASE_URL="postgres://user:pass@host:5432/dbname" \
  --env CASIBASE_SSO_PROVIDER="placeholder" \
  --env CASIBASE_ADMIN_PASSWORD="secure-password"

How to use

Casibase is an AI knowledge base and MCP (model-context-protocol) management platform with an admin UI, user management, and single sign-on (SSO). It orchestrates runtime connections to multiple AI models and providers (e.g., OpenAI, Claude, Llama, Ollama, HuggingFace) and offers a web-based Admin UI to configure agents, permissions, and integrations, along with a chat-driven interface for end users. After deployment, you can access the Admin UI to manage users, roles, and SSO settings, and use the Chat Bot UI to interact with configured models via the MCP protocol. The platform exposes APIs for managing agents, contexts, and model prompts, enabling you to build enterprise workflows and agent-to-agent (A2A) interactions within your organization.

How to install

Prerequisites:

  • Docker installed on your host (Docker Engine 20.10+ or newer)
  • Optional: docker-compose if you prefer compose-based deployment

Installation steps:

  1. Pull or run the Casibase image via Docker:
docker pull casbin/casibase
  1. Run the container (example with basic env vars and port mapping):
docker run -d \
  -p 8000:80 \
  --name casibase \
  -e CASIBASE_ADMIN_EMAIL=admin@example.com \
  -e CASIBASE_ADMIN_PASSWORD=secure-password \
  -e CASIBASE_SSO_PROVIDER=placeholder \
  -e CASIBASE_DATABASE_URL=postgres://user:pass@host:5432/dbname \
  casbin/casibase
  1. Verify startup and access the Admin UI and Chat UI:
  1. (Optional) If you prefer docker-compose, create a docker-compose.yml similar to:
version: '3'
services:
  casibase:
    image: casbin/casibase
    ports:
      - "8000:80"
    environment:
      - CASIBASE_ADMIN_EMAIL=admin@example.com
      - CASIBASE_ADMIN_PASSWORD=secure-password
      - CASIBASE_SSO_PROVIDER=placeholder
      - CASIBASE_DATABASE_URL=postgres://user:pass@host:5432/dbname
  1. Configure models and MCP connections via the Admin UI, then test end-to-end by creating a user, logging in, and invoking model-context flows through the MCP endpoints.

Additional notes

Tips and common issues:

  • Ensure the database URL is reachable from the container (e.g., run the DB in a separate container or a reachable external DB).
  • If the Admin UI does not load, check container logs with: docker logs casibase
  • For SSO, configure your identity provider settings in the Admin UI and ensure the callback URL is whitelisted in your provider.
  • Memory and CPU limits may affect model loading; allocate sufficient resources for the number and size of models you plan to use.
  • If you need to upgrade the image, pull the latest casbin/casibase and recreate the container while preserving config in a mounted volume if applicable.

Related MCP Servers

Sponsor this space

Reach thousands of developers