whereismyai
Webwright
Webwright is a browser agent framework that writes and executes Python scripts to perform long-horizon web tasks using a terminal-based interaction loop.
Webwright provides a terminal environment for LLMs to launch and control browser sessions via generated Python code. It treats the browser as a disposable environment, prioritizing local code artifacts, logs, and screenshots over persistent browser sessions. The framework supports multiple backends, including OpenAI, Anthropic, and OpenRouter, and functions as a plugin for agents like Claude Code, Codex, OpenClaw, and Hermes. Its Skill Factory capability converts successful task trajectories into reusable, parameterized scripts that execute without model intervention. The system maintains a transparent loop, exposing the agent's actions through readable Python scripts.
Webwright at a glance
Key features
- Generates Python scripts for browser interaction
- Integrates with Claude Code, Codex, OpenClaw, and Hermes
- Stores browser task history as local code and logs
- Distills solved tasks into reusable parameterized skills
- Supports OpenAI, Anthropic, and OpenRouter backends
- Operates within a Playwright environment
Use it for
- Automating long-horizon web tasks
- Creating reusable automation scripts
- Developing and debugging browser-based software agents
- Converting complex web interactions into executable programs
Alternatives to Webwright
- BrowserOS — Runs entirely locally and lets AI agents use real user logins to complete tasks.
- Playwright Skill — Generates and runs real, persistent Playwright code instead of simple scripts.
- MyAgent — It features built-in permission controls, long-term memory, and skill self-evolution.
- WebBrain — Runs locally or in the cloud with your choice of LLM and zero telemetry.
- Runway — Control Claude Code and Codex sessions on local or cloud machines from any phone.
- Dev Browser — Persistent browser tabs keep state across scripts so agents do not restart.
- Atomic Agents — Build predictable, modular AI applications with consistent, reliable outputs.
- Cheshire Cat — Build and extend custom artificial intelligence agents using pure Python code.
- Alook — Connects existing local coding agents to team communication channels.
- browser-control — Shell-native design with stable element refs and no LLM or framework requirement.
- Harness2 — It recursively refines agent execution structures using parallel and sequential search.
- Steel — Run scalable cloud browser sessions with a simple developer API.
All Webwright alternatives