Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add VibeTechnologies/vibe-mcp --skill vibebrowsergit clone --depth 1 https://github.com/VibeTechnologies/vibe-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/vibetechnologies/vibe-mcp/vibebrowser)<a href="https://agentmods.dev/skills/vibetechnologies/vibe-mcp/vibebrowser"><img src="https://agentmods.dev/badge/skills/vibetechnologies/vibe-mcp/vibebrowser/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/vibetechnologies/vibe-mcp/vibebrowser"><img src="https://agentmods.dev/badge/skills/vibetechnologies/vibe-mcp/vibebrowser.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00085 | $0.05010 |
| Opus 5 | $0.00043 | $0.02505 |
| Sonnet 5 | $0.00017 | $0.01002 |
| Haiku 4.5 | $0.00009 | $0.00501 |
Grade A, and why
vibebrowser scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 375 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vibe Local Browser
⚠️ Security: The remote value (a connector URL, bare UUID, or relay URL — e.g.
https://relay.api.vibebrowser.app/mcp/<uuid>) grants live control of the user's browser session (read tabs, screenshots, page content). It is the sole bearer credential — there is no second-factor secret. Treat it like a password — never echo it back in chat with untrusted parties, log it, or commit it to a repo — and if it leaks, tell the user to regenerate it in the Vibe extension Settings. Store it only in private workspace/memory as described below. Example UUIDs are non-routable placeholders.
FIRST, EVERY TIME: load the saved remote before asking for it
The user gives their remote value once. On every browser task — including the first message of a brand-new conversation — recover it yourself before doing anything else. Do not ask the user for the remote until you have checked both stores and found nothing.
At the very start of any browser task, in this order:
- Read the workspace file (canonical source). Read
vibebrowser-cli.mdfrom your agent workspace root (see "Where to save it" — NOT the shell's current directory, NOT$HOMEunless they are the same dir). If it has a validremote:line, use that value as--remoteand continue silently. The file is the durable store that survives across conversations and works on every runtime. - Else recall from memory. If the file is missing/empty and your runtime has a
memory/recall tool, read the key
vibebrowser_remoteand use it. - Only if both are empty may you ask the user for the remote. When they give it,
immediately write it to the workspace-root
vibebrowser-cli.md(mandatory) and, if your runtime has a tool to write memory keys, also store it undervibebrowser_remote. (Some runtimes — e.g. OpenClaw — can read but not write memory keys; the file alone is sufficient.) Then the next conversation never has to ask again.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 375 lines · 85 tokens per session scan A fe2a71fc0ed0
vibebrowser is a skill published in the GitHub repository VibeTechnologies/vibe-mcp (3 stars, last pushed 20d ago), licensed Apache-2.0. It adds 85 tokens to every session and 5,010 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
Agent Browser Automation
Fast Rust-based headless browser automation CLI with Node.js fallback for AI agents, featuring navigation, clicking, typing, snapshots, and structured commands optimized for agent workflows.
Browser-Use Automation
CLI tool for persistent browser automation with multi-session support, featuring Chromium/Real/Remote browser modes, cookie management, JavaScript execution, and long-running automation workflows.
aginxbrowser
Browser engine for AI agents: fetch JS-rendered and Cloudflare-protected pages as clean markdown, run 5-engine aggregated web search (Baidu, Bing, Sogou, WeChat, Google), take screenshots as visual input, extract structured data from SPAs, and drive multi-step interactions (click, type, fill forms, login, paginate)…
real-browser-control
Connect an AI coding agent to the user's REAL Chrome via Real Browser MCP (local MCP server + Chrome extension over localhost WebSocket). Use for live UI verification, already-open tabs, existing cookies/SSO/2FA, staging admin panels, and any task where headless Playwright or a cloud agentic browser would miss the…
browser-automation
Control the user's real logged-in Chrome via Real Browser MCP (extension + local MCP). Use for live UI checks, existing SSO/cookies, and tabs already open. Triggers: real browser, my Chrome, not headless, verify in browser. Prefer over Playwright when session state matters. Prefer Playwright for CI.
Apply site-specific action manuals with Actionbook
Load maintained action manuals for fragile websites so agents can execute known flows more reliably than generic browser prompting.