Built-in skill

Companion Face

Creativev0.1.0

Give an agent control of its own companion face on desktop or Android. It can express gestures and moods, speak with lip-sync, compose idle cycles, and invent validated facial expressions. AI-authored content is badged in the Studio.

Android — supported
iPhone — not available
Mac — supported

Get OpenAlly

Grant model

How granting works

A skill is a permission you hand to an agent — not a silent preload.

1Pick it in SkillsAll 24 built-in skills live in one list. Companion Face is there on day one — nothing to install.
2Grant it per agentA skill is a permission, not a preload. Each agent only carries the skills you've granted it, and you can take one back any time.
3It reaches only what's declaredThis one declares no access at all — the ledger below is the receipt.

Skills

Video GenerationGenerate videos from a text prompt — and optional reference images, video, or audio — then have your AI post the finished clip back into the chat. Many providers sit behind one surface (Google Veo, OpenAI Sora, Runway, fal, OpenRouter, MiniMax, DashScope Wan, Seedance, PixVerse, and more); connect a key and pick a default model in Video Providers. Long renders run in the background: the AI starts the job, tells you it's working, and sends the video when it's ready. Video Generation is a per-agent skill — assign it to the agents that should have it (the Executive Assistant always does).

Companion FaceGive an agent control of its own companion face on desktop or Android. It can express gestures and moods, speak with lip-sync, compose idle cycles, and invent validated facial expressions. AI-authored content is badged in the Studio.

Agent AdminBuild and run your own team of sub-agents — create a focused agent for a recurring job, give it a persona and an FAQ, grant it skills and business access, test it, and clean it up.

App AutomationsBuild and run reusable app automations — your AI opens an app and taps, types, and scrolls through a flow for you (post to LinkedIn, search & play on YouTube, summarise Instagram activity, book an Uber up to payment, and more). Give it a topic and it writes the post — you pick from a few drafts before it publishes. Describe the flow once and your AI walks it on the real screen to learn the steps, then replays them fast and deterministically; it self-heals a step if the app changes, and hands off to you at logins and payments. Manage them from any chat. Schedulable. On-device, with your confirmation.

App BuilderBuild, edit, and run on-device React web apps & websites — scaffold a project, edit its source with read-before-write safety, compile it with esbuild, and style it from 14 curated design presets with preloaded fonts.

App GuideKnows OpenAlly inside-out — how to do anything in the app, the skills system, providers, Aster, privacy, support, and how to follow OpenAlly for feature updates. In in-app chat it can drop tappable buttons that open the right screen; in a messaging channel it gives written step-by-step directions.

Companion Face

Companion FaceBuilt-in skill

NetworkNo network access

FilesNo file access

MemoryNo memory access

CredentialsNone stored

Grant to an agent

What the agent is told

Control your own companion face: express gestures and moods, speak with lip-sync, and create expressions and idle cycles.

Prompts

Try it with

The quick actions this skill puts in front of you in the app.

Wave helloWave at me from your companion face and say hello out loud.
Invent an expressionInvent a brand-new facial expression for yourself — something playful — and add it to your face.
Calm morning cycleCreate a calm morning idle cycle for your face and switch to it.

Capability ledger

What it can reach

Straight from the manifest. 19 of 24 skills declare no access — this is one of them.

Network: No network access
Files: No file access
Memory: No memory access
Credentials: None stored
Media: No media access
Browser: No browser

More Creative

Video Generation
Generate videos from a text prompt — and optional reference images, video, or audio — then have your AI post the finished clip back into the chat. Many providers sit behind one surface (Google Veo, OpenAI Sora, Runway, fal, OpenRouter, MiniMax, DashScope Wan, Seedance, PixVerse, and more); connect a key and pick a default model in Video Providers. Long renders run in the background: the AI starts the job, tells you it's working, and sends the video when it's ready. Video Generation is a per-agent skill — assign it to the agents that should have it (the Executive Assistant always does).
- Android — supported
- iPhone — supported
- Mac — supported

Get Companion Face

It ships inside OpenAlly — one install, the whole catalogue.