Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@dokobot/cli) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

@dokobot/cli

Extract structured data from any web page using AI. Powered by Dokobot.

Install

npm i -g @dokobot/cli

Quick Start

# Extract data from a web page
dokobot crawl create \
  --url https://news.ycombinator.com \
  --fields '[{"name":"title","type":"string","required":true},{"name":"points","type":"number","required":true}]' \
  --format csv

# Check task progress
dokobot crawl status <taskId>

# Download results
dokobot crawl result <taskId> --output data.csv

Requires a connected doko (browser extension or dokobot doko connect). See Doko below.

Commands

`crawl`

Extract structured data from web pages using AI.

`crawl create`

Create a new crawl task.

dokobot crawl create \
  --url https://example.com \
  --fields '[{"name":"product","type":"string","required":true},{"name":"price","type":"number","required":true}]' \
  --format csv

Option	Description
`--url <urls...>`	Target URLs (multiple allowed)
`--fields <json>`	Extraction field definitions
`--format <type>`	Output format: `csv`, `json`, or `markdown` (default: `csv`)
`--max-items <n>`	Max items per page (default: `50`)
`--device <id>`	Specific doko device ID

All options can also be provided interactively when omitted.

`crawl list`

dokobot crawl list                          # List recent tasks
dokobot crawl list --status running --limit 5

`crawl status <taskId>`

Show task details, progress, and per-page extraction results.

`crawl result <taskId>`

Download extracted data.

dokobot crawl result <taskId>                    # Print to stdout
dokobot crawl result <taskId> --output data.csv  # Save to file

`crawl cancel <taskId>`

Cancel a running task.

`doko`

Manage doko devices — connect your Chrome browser so AI agents can access real web pages, including SPAs, login-gated sites, and dynamic content.

`doko connect`

dokobot doko connect                              # Interactive setup
dokobot doko connect --name "Work Chrome"         # Name your doko
dokobot doko connect --browser-url http://127.0.0.1:9222  # Custom Chrome instance
dokobot doko connect --device-id <id>             # Reconnect an existing doko

Prerequisites: Chrome with remote debugging enabled:

Open chrome://inspect/#remote-debugging
Check "Allow remote debugging for this browser instance"

`doko list`

List all registered doko devices on this machine.

`doko remove <id>`

Remove a registered doko device.

Other commands

dokobot config      # Configure API key and server URL
dokobot update      # Check for updates

Global Options

Option	Description
`--api-key <key>`	API key (or set `DOKO_API_KEY` env)
`--server <url>`	Server URL (default: `https://dokobot.ai`)
`--verbose`	Enable verbose logging

License

MIT