);
}
```
--------------------------------
### Initialize Greenfield HAPI Project
Source: https://github.com/la-rebelion/mcp-docs/blob/main/docs/2-getting-started/hello-world.mdx
Use the 'hapi init' command to create a new HAPI project based on an OpenAPI schema. The 'example' parameter specifies the name of your project, derived from the schema file name.
```sh
# Check available API Specs
hapi list
# GREENFIELD: Create a new project with the OpenAPI schema
hapi init example
```
--------------------------------
### Integrate WorkOS Authentication with HAPI MCP
Source: https://context7.com/la-rebelion/mcp-docs/llms.txt
Integrates enterprise authentication using WorkOS AuthKit and Dynamic Client Registration. Security is derived from the OpenAPI specification. This example shows downloading demo specs and starting the server.
```bash
# Download a demo spec and a security-extension overlay
curl -o demo.yaml https://docs.mcp.com.ai/apis/openapi/demo.yaml
curl -o api.yaml https://docs.mcp.com.ai/apis/openapi/api-to-mcp.yaml
```
```bash
# Start the server with WorkOS auth enabled
hapi serve demo -f "api.yaml" --headless
```
```yaml
x-hapi:
security:
- oauth2_authorization_code:
client_id: "YOUR_WORKOS_CLIENT_ID"
client_secret: "YOUR_WORKOS_CLIENT_SECRET"
cors:
allowedOrigins:
- "https://your-mcp-client.example.com"
```
--------------------------------
### Start and Inspect LinkedIn MCP Server
Source: https://context7.com/la-rebelion/mcp-docs/llms.txt
Start the LinkedIn MCP server locally and inspect it using the MCP Inspector. This allows for testing and debugging the wrapped API.
```bash
# 3. Start the LinkedIn MCP Server
hapi serve linkedin --headless --port 3030 --url https://api.linkedin.com
# 4. Inspect with MCP Inspector
hapi serve linkedin --headless --port 3030 --url https://api.linkedin.com \
| bunx @modelcontextprotocol/inspector
```
--------------------------------
### RunMCP Docker Compose Example (Dev)
Source: https://github.com/la-rebelion/mcp-docs/blob/main/docs/3-components/runmcp/deployment-models.md
A basic Docker Compose configuration for running a single RunMCP gateway instance. Suitable for development and testing environments.
```yaml
version: '3.8'
services:
runmcp:
image: larebelion/runmcp:latest
ports:
- "8080:8080"
volumes:
- ./config:/app/config
```
--------------------------------
### Run HAPI MCP Server in Air-Gapped Environment
Source: https://github.com/la-rebelion/mcp-docs/blob/main/docs/8-deployment/docker.md
Starts a HAPI MCP server in an air-gapped environment using a locally loaded image and a specified offline API URL.
```sh
docker run --name hapi-offline -d --rm \
-p 3030:3030 \
-v ~/.hapi:/app/.hapi \
hapimcp/hapi-cli:latest serve petstore \
--port 3030 \
--headless \
--url "https://petstore3.air-gapped.local/api/v3"
```
--------------------------------
### API Handler Example
Source: https://github.com/la-rebelion/mcp-docs/blob/main/docs/4-examples/agent-integration.md
Example of an API handler that processes agent actions. This handler is designed to work with the OpenAPI schema defined for agent interactions.
```javascript
app.post('/agent/act', (req, res) => {
const { action } = req.body;
// Implement custom logic or call plugins
res.json({ result: `Action ${action} performed!` });
});
```
--------------------------------
### Serve HAPI MCP Project Locally or Remotely
Source: https://github.com/la-rebelion/mcp-docs/blob/main/docs/3-components/hapi-server/hapi-cli.md
Starts a HAPI MCP server for a local project or a remote OpenAPI URL. Options can be appended for customization.
```bash
# local file
hapi serve