Vibe Browser CLI for any agent runtime
One command surface for OpenClaw, custom operators, and MCP-connected agents. Use your real logged-in browser session locally or through remote relay.
Real browser session
Operate against the tabs, cookies, and logins that already exist in your daily browser profile.
Local or relay mode
Keep one command shape and switch between local and remote UUID routing as your deployment changes.
CLI + MCP in one package
Use vibebrowser-cli for shell flows and vibebrowser-mcp when orchestration grows.
Quickstart
CLI commands that stay explicit
Start with simple browser verbs, then move to the OpenClaw bridge command if your runtime expects it.
Inspect browser state
Confirm a live connection and capture machine-readable page state before acting.
npx -y --package @vibebrowser/mcp@latest vibebrowser-cli --json snapshotExecute explicit browser actions
Use open, click, and type verbs against your real session without launching a disposable browser.
npx -y --package @vibebrowser/mcp@latest vibebrowser-cli click A12Run the same flow remotely
Swap local control for UUID relay access when the browser and agent are on different machines.
npx -y --package @vibebrowser/mcp@latest vibebrowser-cli --remote YOUR_UUID --json statusFor OpenClaw-style bridge routing, run:
npx -y --package @vibebrowser/mcp@latest vibebrowser-mcp openclaw --remote YOUR_UUIDNeed richer tool orchestration? Use the same package via /mcp setup without changing the execution layer.
Built for more than one agent ecosystem
OpenClaw workflows
Keep OpenClaw-style command flows, but execute against your real logged-in browser state.
Custom agent runners
Use the CLI in shell scripts, task runners, and operator runbooks with deterministic steps.
MCP-connected coding agents
Escalate from CLI commands to MCP tooling with the same published package when needed.