> ## Documentation Index > Fetch the complete documentation index at: https://docs.bentolabs.ai/llms.txt > Use this file to discover all available pages before exploring further. # Agent > Drive Bento's in-product AI agent from the terminal. Bento ships an in-product AI agent, the Agent page in the dashboard, that investigates your runs, issues, and signals for you. Ask it a question and it digs through the evidence and answers. The `agent` group drives that same agent from your terminal: open a conversation, send a message, then read or stream the reply. The agent already knows your workspace, so you don't hand it any context. You talk to it in a conversation, the thread that holds a title and its messages. Each message kicks off a turn, one run of the agent for that message. A turn does real work, reading your runs, issues, and signals before it answers, so it runs in the background: you send the message, get a `turn_id` back right away, then read or stream that turn as it finishes. Each entry in the thread is a message, yours or the agent's reply. Sometimes a turn proposes a confirmation, an action the agent wants to take that needs your `approved` or `rejected` sign-off. ## Commands ```bash theme={null} bentolabs agent --help ``` | Command | What it does | | ------------------------------------------ | ---------------------------------------------------------- | | `agent list-conversations` | List your conversations. Page with `--limit` / `--offset`. | | `agent create-conversation` | Start a new conversation. | | `agent get-conversation ` | Open one conversation with its messages. | | `agent enqueue-chat` | Send a message. Returns the `turn_id` to read next. | | `agent get-turn ` | Read a turn and its events once. | | `agent stream-turn ` | Stream a turn's events as they happen. | | `agent confirm-action-route` | Approve or reject an action the agent proposed. | ## Talk to the agent Start a conversation. `title` is optional; leave it off and the agent names it: ```bash theme={null} bentolabs agent create-conversation --data '{"title": "Prod incident triage"}' ``` Send a message. `message` is required; pass the `conversation_id` to keep the thread going. Skip the first step entirely by calling `enqueue-chat` with only a `message`, and the agent starts a fresh conversation and returns both ids. The response carries the `turn_id`: ```bash theme={null} bentolabs agent enqueue-chat --data '{"conversation_id": "", "message": "Why did our checkout agent start failing today?"}' ``` Read the turn once it has run in the background: ```bash theme={null} bentolabs agent get-turn ``` Or stream its events live as the agent works. Resume from where you left off with `--after-seq`: ```bash theme={null} bentolabs agent stream-turn --after-seq 42 ``` ## Review the thread List your conversations, newest activity first: ```bash theme={null} bentolabs agent list-conversations --limit 20 --output table ``` Open one to read its full message history: ```bash theme={null} bentolabs agent get-conversation ``` ## Confirm an action When the agent proposes an action that needs sign-off, answer with a decision. `conversation_id` and `action` are required; `decision` is `approved` or `rejected`: ```bash theme={null} bentolabs agent confirm-action-route --data '{"conversation_id": "", "action": "", "decision": "approved"}' ``` ## See also The tracked problems the agent investigates. The detectors behind those problems. What the CLI is and how the groups fit together.