Television

Television is a visual interface for personal agents that provides a dedicated space for creating, arranging, and reusing artifacts.

Television functions as a visual interface for personal agents, allowing users to move work out of chat windows and into organized, persistent environments. The product allows users and their agents to create, arrange, and reuse artifacts such as simple cards, interactive views, or external web pages. Users organize these artifacts into channels to collect and return to specific tasks. It integrates with any agent harness capable of running commands and writing files. Television provides skills, which are reusable instructions for creating and sharing these interfaces. The tool is designed for users who want to maintain a visual record of agent outputs over time.

Television at a glance

Best for: Software engineer, Product manager
Runs on: Web
Pricing: Free
Released: October 2026
Category: AI agent tool
Website: television.run

Key features

Artifact creation for pinning important agent information
Persistent channels for organizing artifacts
Interactive views and web page integration
Reusable skill definitions for interface creation
Agent harness integration via command execution and file writing

Use it for

Collecting and arranging project artifacts
Revisiting specific agent work over time
Sharing adaptable interfaces with others

Alternatives to Television

The New Beings — Watch AI agents perform a live talk show and connect your own agents.
Agent Office 3d — Visualizes AI coding agent states and terminal outputs in a 3D office.
Adyen — Share AI agent resources across coding tools and repositories easily.
Foldspace Product Agent — It acts inside your live product through APIs and permissions.
Page Agent — Integrates directly into web pages without browser extensions or headless browsers.
Yao — Self-hosted agent platform with task tracking on centralized boards.
Agenta — Build and automate AI agents that run on your own infrastructure or the cloud.
Open Interface — It uses visual screenshots to self-correct its actions while navigating GUIs.
MS-Agent — It supports multimodal inputs, autonomous scheduling, and modular tool integration.
campfire — It gives humans and AI agents a shared workspace to collaborate.
video2ctx — Provides video context with timestamped citations and uses 95% fewer tokens than transcripts.
Browser Use Web UI — It supports custom browser integration and persistent sessions.

All Television alternatives