Ottonomous

Showing 25 of 39 from the live probe.

• Auth

  Manage authentication — check status, login, 2FA, set token, or logout.

• Stores

  Manage stores, users, groups, invitations, locations, departments, hours, and categories.

• Collections

  Manage collection templates, fields, field groups, directories, and lifecycle states.

• Entries

  Search, create, update, delete, and manage collection entries and their assets.

• Assets

  Browse, upload, move, and delete files and folders in the asset library.

• Integrations

  Manage external integrations such as FTP, SFTP, and API connections.

• Automations

  Manage automation apps, actions, triggers, connections, and execute one-off actions.

• Workflows

  Manage workflows, triggers, actions, folders, and execution logs.

• Websites

  Manage websites, pages, components, and filters for a store.

• Api Reference

  Get API specs, environment config, or search endpoints for Otto Build.

• Metrics

  Manage metrics, dashboards, and evaluate metric/dashboard data.

• Web Fetch

  Fetch a URL and return its content as clean readable text. Uses a plain HTTP GET first; falls back to headless Chromium if the page appears to require JavaScript rendering.

• Web Scrape

  Launch a headless browser, load a URL, and return structured visible content: page text, links, and interactive elements (buttons, inputs, selects). No AI processing — returns raw content for the calling agent to reason over.

• Computer Status

  Check if the Ottonomous desktop app is running. Returns platform, version, and current target. Call first to confirm availability before any other computer_* tool.

• Computer Set Target

  Start a browser automation session. Opens a new isolated window in the user's existing browser (inheriting their cookies and logins) and applies stealth patches. The best available browser is auto-detected (Chrome > Arc > Brave > Edge) — only pass browser to override. For native (non-browser) apps, pass app only.

• Computer Orient

  Analyze the current browser page in one pass and return a complete, structured snapshot. Always call this after navigation or form submission — it tells you everything you need to plan the next action. Returns: - pageType: 'form' | 'list' | 'captcha' | 'confirmation' | 'login' | 'error' | 'unknown' - textInputs / comboboxes / checkboxes / fileInputs — all visible fields with CSS selectors, labels, and current values - comboboxes.probedOptions — for native <select> elements, the available option

• Computer Act

  Execute an ordered sequence of browser actions in a single round-trip. Use after computer_orient to drive the page. Each step is an object with a 'type' field. Step types: PRIMARY (use these for web forms): - execute { type, plan: [{action, ...}] } Run a complete interaction plan inside the page via the Otto runtime. Zero per-action round-trip cost — the entire plan executes locally and returns one result array. PREFER this over individual click/type/select steps whenever fill

• Computer Fill Form

  Fill an entire visible form in one call. Pass all the data you want to fill at once — the desktop runtime handles everything: text inputs, React dropdowns, native selects, and checkboxes. HOW IT WORKS: 1. Discovers all visible fields on the page, scoped to the active form/dialog 2. Matches each data key to the best-scoring field by label (aria-label, placeholder, nearby text) 3. Fills text inputs directly with React-aware events 4. Drives React dropdowns with trusted CDP mouse clicks + Mutation

• Computer Shell

  Run a shell command on the user's desktop machine and return stdout + stderr. Use this for file operations, downloads, and system tasks that don't require browser interaction. Common uses: - Download a file before uploading: curl -o /tmp/photo.jpg 'https://...' - List or move files: ls ~/Downloads, mv file.pdf /tmp/ - Read a local file: cat /Users/alice/vehicles.csv - Check what's at a path: file /tmp/photo.jpg Output is capped at 10 MB. Default timeout is 30 seconds; pass timeout (ms) to exte

• Computer Screenshot

  Capture the current window (or full screen if no target set) as a base64 PNG. Use for visual verification, CAPTCHA solving, or interacting with native apps where DOM tools can't reach.

• App Stripe

  Stripe — Third-party integration. Use overview for capabilities, search_docs to find endpoints, describe for endpoint details, execute to make API calls.

• App Google Ads

  Google ads — Third-party integration. Use overview for capabilities, search_docs to find endpoints, describe for endpoint details, execute to make API calls.

• App Google Calendar

  Google calendar — Third-party integration. Use overview for capabilities, search_docs to find endpoints, describe for endpoint details, execute to make API calls.

• App Google Drive

  Google drive — Third-party integration. Use overview for capabilities, search_docs to find endpoints, describe for endpoint details, execute to make API calls.

• App Google Search Console

  Google search console — Third-party integration. Use overview for capabilities, search_docs to find endpoints, describe for endpoint details, execute to make API calls.