GH Repository · agentrhq
webcmd
The Browser Infra that learns and cuts token spend by upto 90%
- stars
- 2,252
- 30-day movement
- +658219/day
- Related entries
- 61
- Connections
- 1
makeTypeScripttypescriptself-learning-aiplaywrightnode/bunbrowser-usedeveloper-toolsai-agentsagent-appclinodebrowser-automation
webcmd is a browser automation infrastructure tool for AI agents, described as self-learning and aimed at cutting token spend by up to 90%. It is a TypeScript CLI built on Playwright and distributed as an agent app in the agentrhq toolchain.
Reach for it when your agents drive a browser and you want to reduce the token cost of that interaction.
Use it to
- Automate browser tasks from AI agents
- Reduce token spend in agent browser loops
- Run Playwright-based automation via CLI
- Build agent apps that need browser control
For Developers building browser-using AI agents
- Role
- agent-app
- Language
- TypeScript
- Licence
- Apache-2.0
- Forks
- 1,527
- Open issues
- 18
- Last push
- 2026-09-11
- Latest release
- webcmd-v0.1.0 · 2026-07-03
- Skills shipped
- 1
topicsbrowser-automationai-agentsplaywrightclitoken-optimizationdeveloper-tools