MachinaOS

Drive the user's real Chrome over raw CDP by writing Python against browser-harness helpers - screenshot, coordinate clicks, JS evaluation, form fill, tabs. For tasks needing full freedom or the user's own logins.

Work with GitHub via the gh CLI — clone repositories, create/list/merge pull requests, create/list issues, and run any other gh command (API calls, workflow runs, releases, repo administration). List operations return parsed JSON.

Deploy sites and apps to Vercel, inspect deployments, stream logs, and manage projects/env/domains via the Vercel CLI. Deploy a directory and get back the live deployment URL; everything else the CLI supports is available through the custom command passthrough.

Use this skill to generate well-branded interfaces and assets for MachinaOS (zeenie.ai), either for production or throwaway prototypes/mocks/etc. Contains essential design guidelines, colors, type, fonts, assets, and UI kit components for prototyping.

Run AI-generated Python in a hard sandbox (Pydantic Monty) with enforced time + memory limits and opt-in capabilities. Use for untrusted code; supports a Python subset.

Interactive browser automation - navigate, click, type, fill forms, take screenshots, get accessibility snapshots. Supports system Chrome/Edge via auto-detection.

Write new files or edit existing files with string replacement.

Read file contents with line numbers and pagination support.

Search the filesystem with ls (list directory), glob (pattern match), or grep (search file contents).

Process payments, manage customers, issue refunds, and handle subscriptions via the Stripe CLI. Pass any Stripe command (customers, charges, payment_intents, refunds, invoices, products, prices, subscriptions) and the tool runs it for you and returns parsed JSON.

How to use the Agent Builder tool's five canvas-mutation operations to inspect and grow your own toolset / skills / teammates / workflows mid-execution

ALWAYS call the wired chat-model tool (anthropic_chat_model / openai_chat_model / gemini_chat_model) at the START of any complex task to get a plan from a stronger model. Also call when stuck and before declaring done. Pass only the `prompt` field — the operator configured `model` and `api_key`. Returns brief tactical advice; you do the work.

General AI assistant with task planning via write_todos, helpful and honest responses.

Create and manage structured task lists for planning and tracking complex multi-step operations.

Search for recent tweets on Twitter/X using keywords, hashtags, mentions, and advanced query operators. Returns rich tweet data with expanded URLs, author info, media, metrics, and referenced tweets.

Execute short-lived shell commands inside the per-workflow workspace. The shell is Nushell (cross-platform, same syntax on Windows/macOS/Linux). External tools on PATH (npm, node, python, git, ...) are available.

Execute Python code for calculations, data processing, and automation. Access to math, json, datetime, collections, and more.

Android device assistant personality. Use when the user wants to interact with their Android phone or tablet in a natural, conversational way.

Manage memory compaction by summarizing conversation history when approaching token limits

Output human-readable text in pretty formats instead of raw markdown or code blocks

Manage conversation memory. Use when user asks to remember something, recall information, view/edit conversation history, or manage short/long-term memory.

Manage sub-agent delegation, handle task completion events, and coordinate multi-agent workflows.

Discover and load tools progressively as needed (reduces token overhead)

Guides the RLM agent to use REPL code execution with recursive LM calls for complex reasoning, decomposition, and multi-step problem solving.

Launch Android applications by package name. Open any installed app programmatically.

Get list of installed Android applications with package names, versions, and metadata.

Control Android audio - get/set volume, mute/unmute for media, ringtone, notification, and call volumes.

Monitor Android device battery status, level, charging state, temperature, and health.

Control Android Bluetooth - enable, disable, get status, and list paired devices.

Control Android camera - get camera info, take photos, and access camera capabilities.

Get Android environmental sensor data - temperature, humidity, pressure, and ambient light level.

Get Android device GPS location - latitude, longitude, accuracy, speed, and provider information.

Get Android motion sensor data - accelerometer, gyroscope, detect motion, shake gestures, and device orientation.

Control Android screen - brightness, wake screen, auto-brightness toggle, and screen timeout settings.

Control Android WiFi - enable, disable, get status, scan for networks, and get current connection info.

Autonomous decision loop with reflection and iteration

Generate Python code instead of sequential tool calls (81-98% token savings)

Handle errors gracefully with retry strategies and fallback patterns

Coordinate multiple tools together for complex multi-step tasks

Execute JavaScript code for calculations, data processing, and JSON manipulation. Full ES6+ support with Node.js runtime.