Usage: kimi [options] [command]

The Starting Point for Next-Gen Agents

Options:
  -V, --version                 output the version number
  -S, --session [id]            Resume a session. With ID: resume that session. Without ID:
                                interactively pick.
  -c, --continue                Continue the previous session for the working directory. (default:
                                false)
  -y, --yolo                    Start in Ask When Needed mode: routine edits and commands run
                                automatically; risky actions, questions, and plans still ask.
                                (default: false)
  --auto                        Start in Never Ask mode: never interrupts you; everything runs and
                                is decided automatically. (default: false)
  -m, --model <model>           LLM model alias to use for this invocation. Defaults to
                                default_model in config.toml.
  -p, --prompt <prompt>         Run one prompt non-interactively and print the response.
  --output-format <format>      Output format for prompt mode. Defaults to text. (choices: "text",
                                "stream-json")
  --skills-dir <dir>            Load skills from this directory instead of auto-discovered user and
                                project directories. Can be repeated. (default: [])
  --agent <name>                Agent profile to start the new session with. Custom profiles are
                                discovered from agent directories or loaded via --agent-file. Cannot
                                be combined with --session/--continue.
  --agent-file <path>           Load an agent definition from a Markdown file and select it for the
                                new session. Cannot be combined with --session/--continue. (default:
                                [])
  --add-dir <dir>               Add an additional workspace directory for this session. Can be
                                repeated. (default: [])
  --plan                        Start in plan mode. (default: false)
  -h, --help                    Show help.

Commands:
  export [options] [sessionId]  Export a session as a ZIP archive.
  fork [options] [sessionId]    Fork a session into a new session.
  provider                      Manage LLM providers non-interactively.
  session                       Manage sessions non-interactively.
  acp [options]                 Run kimi-code as an Agent Client Protocol (ACP) server over stdio.
  web [options]                 Run the local Kimi server and open the web UI.
  server                        Deprecated — use `kimi web` instead.
  rc|remote [options]           Run the local Kimi server and open the web UI through Remote
                                Control.
  login [options]               Authenticate with Kimi Code CLI via the device-code flow.
  doctor                        Validate Kimi Code configuration files.
  vis [options] [sessionId]     Launch the session visualizer in your browser.
  migrate [options]             Migrate data from a legacy kimi-cli installation into kimi-code.
  upgrade|update                Upgrade Kimi Code to the latest version.

Documentation:        https://moonshotai.github.io/kimi-code/
