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 WingedGuardian/GENesis-AGI --skill stealth-browsergit clone --depth 1 https://github.com/WingedGuardian/GENesis-AGIWrote 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/wingedguardian/genesis-agi/stealth-browser)<a href="https://agentmods.dev/skills/wingedguardian/genesis-agi/stealth-browser"><img src="https://agentmods.dev/badge/skills/wingedguardian/genesis-agi/stealth-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/wingedguardian/genesis-agi/stealth-browser"><img src="https://agentmods.dev/badge/skills/wingedguardian/genesis-agi/stealth-browser.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00045 | $0.03288 |
| Opus 5.5 | $0.00018 | $0.01315 |
| Sonnet 5.5 | $0.00009 | $0.00658 |
| Haiku 4.5 | $0.00005 | $0.00329 |
Grade A, and why
stealth-browser 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 13d 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 — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Stealth Browser
Behaviour rules for sites that try to detect automation. Read the
browser-automation skill first: layers, timeouts, safety gates,
verify-after-act, overlays and closing browsers live there and are not repeated
here.
Bot detection scores the environment (fingerprint, IP, how the browser is driven) and the behaviour. The browser engine and the tools cover most of both. Your job is the part only a planner can do: what to visit, in what order, what not to touch, and when to stop.
What the tools already do
Do not re-implement any of this.
| Behaviour | Where | Detail |
|---|---|---|
| Fingerprint | Camoufox engine | Firefox-based, spoofed at engine level |
Delay before browser_click, browser_fill, browser_upload |
all layers below | see Timing |
| Per-keystroke typing | Camoufox, remote CDP | field cleared first; key hold log-normal, median 86 ms (clamped 30-200 ms, p95 ~153 ms); 50-200 ms between keys; 5% of gaps are 0.3-1 s pauses. Chromium and TinyFish fill atomically |
| Cursor | Camoufox | scrolls the target into view, humanize=2.5 cursor trail to a point in the central 60% of the element that hit-tests as the element (when none does, Playwright picks the point), 50-200 ms dwell, 40-120 ms press through Playwright's click, which hit-tests again. A cover the hit test finds fails as Click blocked naming it; a styled checkbox or radio is clicked through its own label. See browser-automation, "Clicking" |
| Turnstile | Camoufox, Chromium | detected and worked on inside browser_navigate (below) |
| Keyboard repeat | browser_press_key |
50-150 ms between repeats, no pre-delay |
Not done for you, and not to be faked: idle cursor jitter, humanized
scrolling, tab visibility changes. _idle_jitter() and _human_scroll() exist
in browser.py with no call site. Do not emulate them with browser_run_js:
script-dispatched events carry isTrusted=false and a synthetic
visibilitychange leaves document.visibilityState at visible, so both are
easier to detect than doing nothing.
What ships with it
4 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.
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.
- 13d ago Changed · +1 lines 55b3afeefa7b
- 18d ago Changed 9ba42b5cc5f6
- 19d ago Changed · -145 lines · +32 tokens per session 21b949996647
- 1mo ago Changed · +19 lines c4c684fc6dca
- 1mo ago First seen · 335 lines · 13 tokens per session scan A 95be940c3576
stealth-browser is a skill published in the GitHub repository WingedGuardian/GENesis-AGI (99 stars, last pushed today), licensed MIT. It adds 45 tokens to every session and 3,288 once invoked, about $0.0002 per session on Opus 5.5. 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
stealth-browser-launch
Launch stealth Chromium with C++ fingerprint patches for anti-bot bypass.
bot-protection
Block automated bots from scraping your catalog, scalping limited inventory, and abusing checkout flows using CAPTCHA and behavioral detection.
skills
Anti-detection browser CLI powered by Camoufox. Browse the web, search, extract data, take screenshots, and crawl pages — all with C++ level fingerprint spoofing that bypasses Cloudflare, Google, and most bot detection systems.
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.
implementing-network-deception-with-honeypots
Deploy and manage network honeypots using OpenCanary, T-Pot, or Cowrie to detect unauthorized access, lateral movement, and attacker reconnaissance.