CI and Automation

CI Mode

Run Qodo Gen CLI in CI environments with the --ci flag:

qodo <command> --ci

The output logs from Qodo Gen CLI will be simplified and fit for CI environments and build pipelines.

Webhook Mode

Run Qodo as an HTTP server to receive POST requests for each command.

qodo --webhook

This flag will make Qodo Gen CLI:

• Expose each of your configured agents as a separate POST endpoint.

• Open a server.

• Listen to POST requests that come in through the URLs.

The POST requests are validated according to the arguments available in the configured agent, and return the same return value that's set in the agent.

POST Example:

POST http://localhost:4000/webhook/explain

Returns results at end of execution.

SSE Mode (Live Streaming Logs)

If you add the parameter sse=true to the URL, a session ID will be returned. That enables you to run a GET request with the session ID.

Running a configured agent like so, you can see all the actions the agent is running behind the scenes, not just the return value.

POST http://localhost:4000/webhook/explain?sse=true

-> returns sessionId

GET http://localhost:4000/webhook/sse/explain?sessionId=<ID>

MCP Mode (Tool Server)

When you run Qodo Gen CLI in MCP mode using the --mcp flag, it transforms into a server that exposes all configured agents (only those with a description) as callable tools.

qodo --mcp

Each of the configured agents becomes accessible through a streamable HTTP URL, allowing external systems to trigger them easily.

As the server processes a request, it can send real-time progress notifications and eventually returns the final result of the execution.

This allows you to seamlessly integrate your agents with other tools or platforms without the need to share API keys, as all execution and access are handled securely through the Qodo Gen CLI server itself.

UI Mode

Use the Qodo web interface:

qodo --ui

The web interface features full interactivity, and the same capabilities as in the terminal.

Available Advanced Commands and Flags

Command	Description
<command-name> --mcp	Run an agent as a service (listening on port 3000)
mcp-list	List available local and remote tools
--ci	Run commands in CI mode
--mcp	Run all configured agents as tools in MCP-server mode
--webhook	Run commands as tools from agent.toml in webhook mode
-h, --help	View all available commands and flags
-l, --log=path	Redirect console output to a file (directed in the path), or to stdout or stderr
--silent	Suppress all console output except the final result. Logs go to /dev/null
--resume=session_id	Resume a task with the given session ID