API & MCP Access

Generate API keys to access Jitera agents, project data, and MCP tools from external applications.

API keys let you access your Jitera project’s agents, data, and MCP tools from external applications. The API is OpenAI-compatible — any tool or SDK that works with the OpenAI API can connect to Jitera by changing the base URL and API key.

Navigating to API Keys

From your project, go to Settings > Development > API keys.

Creating an API Key

Only project Owners and Admins can create API keys. All project members can view existing keys.

1. Click + New key

   

2. Enter a descriptive name for the key (e.g., “Claude Desktop”, “CI Pipeline”)

3. Click Create

4. The dialog shows three items — copy all of them immediately:

   • API Secret Key — The full authentication token (shown only once)
   • API Base URL — The endpoint for API requests
   • MCP JSON — A ready-to-use MCP server configuration
   

The API Secret Key is displayed only once at creation. Store it securely. If lost, create a new key.

Using with MCP Clients

The MCP JSON configuration can be copied directly into MCP-compatible tools. For example, to use Jitera’s agents and tools in Claude Desktop or Cursor:

1. Create an API key as described above
2. Copy the MCP JSON from the dialog
3. Paste it into the MCP server configuration of your external tool
4. The external tool can now access your Jitera agents’ tools via MCP

The MCP JSON has the following format:

{
  "mcpServers": {
    "jitera-mcp": {
      "url": "https://mcp.jitera.ai",
      "headers": {
        "Authorization": "Bearer <your-api-secret-key>"
      }
    }
  }
}

Alternative URL: https://proxy-us.jitera.ai/gateway/boost

Enterprise customers with a Japan server deployment use different endpoints. If your organization has opted for a Japan-based server, replace the URL with one of the following:

• https://gateway-proxy.jitera.app/gateway/boost
• https://ai.jitera.app

Using as an OpenAI-Compatible API

The Jitera API follows the OpenAI API specification. You can use any OpenAI-compatible SDK or HTTP client by pointing it to the Jitera API Base URL.

Authentication

All API requests require the Authorization header:

Authorization: Bearer <your-api-secret-key>

Available Endpoints

Method	Endpoint	Description
GET	/v1/models	List available models
GET	/v1/models/{model_id}	Get a specific model
POST	/v1/chat/completions	Chat completion (streaming and non-streaming)
POST	/v1/files	Upload a file
GET	/v1/files	List files
GET	/v1/files/{file_id}	Get file metadata
DELETE	/v1/files/{file_id}	Delete a file
GET	/v1/files/{file_id}/content	Download file content

Chat Completions

POST /v1/chat/completions

Send a chat request to a Jitera agent. The request and response follow the OpenAI Chat Completions format.

Request:

{
  "model": "jitera/chat:latest",
  "messages": [
    { "role": "user", "content": "Hello!" }
  ],
  "stream": false
}

Response:

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1711700000,
  "model": "jitera/chat:latest",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you today?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 12,
    "total_tokens": 22
  }
}

Streaming

Set "stream": true to receive the response as Server-Sent Events (SSE). Each event is a JSON chunk in the data: field:

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant","content":"Hello"},"finish_reason":null}]}

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"!"},"finish_reason":null}]}

data: {"id":"chatcmpl-abc123","object":"chat.completion.chunk","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: [DONE]

Models

Use the model field to select which agent handles the request:

Model ID	Description
jitera/chat:latest	Default Jitera chat agent
jitera/<agent_id>	A specific custom agent (use the agent’s ID)

Call GET /v1/models to list all available models for your project.

Example with OpenAI SDK (Python)

from openai import OpenAI

client = OpenAI(
    api_key="<your-api-secret-key>",
    base_url="<your-api-base-url>/v1",
)

response = client.chat.completions.create(
    model="jitera/chat:latest",
    messages=[
        {"role": "user", "content": "Summarize our project requirements"}
    ],
)

print(response.choices[0].message.content)

Example with cURL

curl <your-api-base-url>/v1/chat/completions \
  -H "Authorization: Bearer <your-api-secret-key>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "jitera/chat:latest",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Managing API Keys

The API keys list shows:

Column	Description
Name	The name you gave the key
Created	When the key was created
Last used	When the key was last used for an API request
Created by	The team member who created the key

To revoke a key, click the delete action next to it. Revoked keys immediately stop working for all API and MCP requests.

Related

• Apps (MCP) — Connecting MCP apps to agents
• Managing Agents — Agent creation and configuration
• Automation — Scheduling agent tasks