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 agentmods add skills/browserbase/stagehand/stagehand-facadenpx skills add browserbase/stagehand --skill stagehand-facadegit clone --depth 1 https://github.com/browserbase/stagehandWhat 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 | $0.00042 | $0.00330 |
| Opus 5 | $0.00021 | $0.00165 |
| Sonnet 5 | $0.00008 | $0.00066 |
| Haiku 4.5 | $0.00004 | $0.00033 |
Grade A, and why
stagehand-facade 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 2d 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.
What it actually says
In fx, the tools are registered as mcp_stagehand_run, mcp_stagehand_snapshot, and
mcp_stagehand_screenshot. Select them with mcp_select_tool using those exact names before
calling them; mcp_search_tools may return no results for this server, so do not rely on it.
Never invent another Stagehand tool name. There is no separate stagehand_navigate,
mcp_stagehand_navigate, or start tool.
You control one persistent browser through exactly three tools:
- snapshot: inspect the active page and hydrate bracketed element IDs.
- run: browse or navigate with JavaScript (for example,
await page.goto(url)) or provide snapshot actions using the Playwright-shaped page API. The first call initializes the browser. - screenshot: inspect the rendered page visually. Prefer viewport JPEG with
{"type":"jpeg","quality":40,"fullPage":false}; inline full-page PNGs may exceed fx's MCP response-frame limit.
Use snapshot actions for simple interactions and run code for multi-step workflows. Pass run exactly one of code or actions; every action uses "op" and "id", never "kind" or "ref". Snapshot IDs are valid only for the latest snapshot of the active page; snapshot again after navigation or stale IDs. Do not launch another browser.
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.
- 2d ago First seen · 22 lines · 42 tokens per session scan A e273519b4f8d
stagehand-facade is a skill published in the GitHub repository browserbase/stagehand (24,098 stars, last pushed 2d ago), licensed MIT. It adds 42 tokens to every session and 330 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
brightdata-web-mcp
Search the web, scrape websites, extract structured data from URLs, and automate browsers using Bright Data's Web MCP. Use when fetching live web content, bypassing blocks/CAPTCHAs, getting product data from Amazon/eBay, social media posts, or when standard requests fail.
agent-browser
Drive a real browser to inspect or interact with a web page or app — navigate, take screenshots, read console and network, fill simple forms — for verification tasks, not unattended automation.
Browser Automation
Automate web browser interactions, scraping, testing, and workflow automation with Puppeteer/Playwright.
fortress-stealth-browser
Use when a web fetch is blocked — Cloudflare, DataDome, PerimeterX, Akamai, a 403/429, a CAPTCHA/"Press & Hold", "Access denied", an empty JavaScript shell, or a page whose data only appears after client-side rendering. Drives a real recompiled-Chromium stealth engine (Fortress) via the Fortress MCP to fetch, extract…
air-sdk
Web intelligence for AI agents. Discover what actions are possible on any website, get verified CSS selectors and step-by-step browser automation plans, extract structured data from URLs, and report outcomes so every agent benefits from shared learning. Covers 25,000+ domains.
air-sdk
Collective web intelligence for browser agents. Discover site capabilities, get CSS selectors, extract structured data, and report outcomes. When one agent learns how to use a website, every agent benefits.