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 appautomaton/webmaton --skill nodriver-browsergit clone --depth 1 https://github.com/appautomaton/webmatonWrote 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/appautomaton/webmaton/nodriver-browser)<a href="https://agentmods.dev/skills/appautomaton/webmaton/nodriver-browser"><img src="https://agentmods.dev/badge/skills/appautomaton/webmaton/nodriver-browser/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/appautomaton/webmaton/nodriver-browser"><img src="https://agentmods.dev/badge/skills/appautomaton/webmaton/nodriver-browser.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high YARA Match · line 165 YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).Fix: Remove the malware payload or compromised file entirely. Investigate how it entered the skill and audit all other artifacts for additional indicators of compromise.
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 | $0.00106 | $0.04011 |
| Opus 5.5 | $0.00042 | $0.01604 |
| Sonnet 5.5 | $0.00021 | $0.00802 |
| Haiku 4.5 | $0.00011 | $0.00401 |
Grade A, and why
nodriver-browser scanned grade A with 1 finding 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **A JSON API endpoint** → use `curl` via Bash, you don't need a browser at all How it starts
The opening of the file, as written. The whole thing — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.
nodriver-browser
A persistent Chrome/Chromium browser that stays alive between Claude's turns. Built on nodriver (CDP-direct, no Selenium, no navigator.webdriver). Every script attaches to the same long-running browser, performs one action, exits — the browser and its tab keep going.
Core invariant: ONE daemon, ONE persistent tab (tabs[0]). Every script reports tabs_open in its output. If you ever see tabs_open > 1, treat it as a real signal that something opened a stray tab — read the warning and act on it.
When to use this skill
- The page needs JavaScript to render (SPA, infinite scroll, lazy load)
- WebFetch is blocked by anti-bot systems (Cloudflare, DataDome, Imperva, hCaptcha)
- The task requires interaction: clicking buttons, filling forms, multi-step flows, dropdown selection
- You need session state across multiple actions (logged-in scraping, multi-page checkout, OAuth flows)
- You need visual proof of a page (screenshot) for debugging or reporting
When NOT to use this skill
- Static HTML that loads fully on first GET → use
WebFetch - One-shot search query → use
WebSearch(or grok-search skill if it exists) - A JSON API endpoint → use
curlvia Bash, you don't need a browser at all - A single quick scrape with no interactivity → consider a one-off Python script, not this skill
This skill spawns a Chromium process (~150-200 MB RAM) that stays alive until you explicitly stop it. Worth it for interactive flows; overkill for one URL.
Quick start
Scripts live in scripts/ next to this SKILL.md — resolve paths relative to the skill root.
Run them directly as executables; the #!/usr/bin/env -S uv run --script
shebang invokes uv and reads the PEP 723 metadata for you. Do not run these
scripts with python, python3, or python -m; that bypasses the shebang,
dependency metadata, and Python version pin. If executable dispatch is
unavailable, use uv run --script scripts/nav.py ....
What ships with it
25 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- lib/runner.py 37 KB runs code
- lib/snapshot.py 4.1 KB runs code
- references/constrained-environments.md 850 B
- scripts/back.py 1.1 KB runs code
- scripts/cleanup.py 1.1 KB runs code
- scripts/click.py 2.5 KB runs code
- scripts/close_tab.py 2.3 KB runs code
- scripts/eval.py 1.4 KB runs code
- scripts/forward.py 1.1 KB runs code
- scripts/hover.py 2.7 KB runs code
- scripts/nav.py 1.4 KB runs code
- scripts/press.py 3.0 KB runs code
- scripts/reload.py 1.1 KB runs code
- scripts/screenshot.py 1.6 KB runs code
- scripts/scroll.py 2.1 KB runs code
- scripts/select.py 3.1 KB runs code
- scripts/snapshot.py 1.6 KB runs code
- scripts/start_daemon.py 1.5 KB runs code
- scripts/state.py 1.3 KB runs code
- scripts/status.py 2.7 KB runs code
- scripts/stop_daemon.py 649 B runs code
- scripts/tabs.py 975 B runs code
- scripts/type.py 2.9 KB runs code
- scripts/upload.py 3.0 KB runs code
- scripts/wait.py 2.3 KB runs code
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.
- 1mo ago First seen · 229 lines · 106 tokens per session scan A 51a26d5b6f06
nodriver-browser is a skill published in the GitHub repository appautomaton/webmaton (20 stars, last pushed 1mo ago), licensed MIT. It adds 106 tokens to every session and 4,011 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work. Also covers hosted cloud browsers for AI agents: the browser-use Python library (Browser(usecloud=True)), Playwright, Puppeteer or any CDP client.
browse
Use the browse CLI for Browserbase browser automation, Browserbase cloud APIs, Browserbase Functions, templates, web fetch/search, diagnostics, and Browse.sh skill discovery/installation. Use when the user asks to navigate pages, inspect browser state, run local or remote browser sessions, manage Browserbase…
opencli-autofix
Automatically fix broken OpenCLI adapters when commands fail. Load this skill when an opencli command fails — it guides you through collecting a trace artifact, patching the adapter, retrying, and filing an upstream GitHub issue after a verified fix. Works with any AI agent.
moli-webfetch
Fetch, inspect, crawl, and capture live, JavaScript-rendered websites with Moli. Use when Codex needs current web content, web research, fact lookup, link following, a bounded crawl, client-rendered or response-gated content, network diagnostics, or a standalone HTML, Markdown, JSON, semantic-tree, viewport or…
pinchtab
Use this skill when a task needs browser automation through PinchTab: open a website, inspect interactive elements, click through flows, fill out forms, scrape page text, reuse a dedicated automation profile with user approval, export screenshots or PDFs, manage multiple browser instances, or fall back to the HTTP API…
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.