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/agentcomputerai/torch/camoufoxnpx skills add AgentComputerAI/torch --skill camoufoxgit clone --depth 1 https://github.com/AgentComputerAI/torchWhat 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 | $0.00091 | $0.02041 |
| Opus 5 | $0.00046 | $0.01020 |
| Sonnet 5 | $0.00018 | $0.00408 |
| Haiku 4.5 | $0.00009 | $0.00204 |
Grade A, and why
camoufox 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.
How it starts
The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Camoufox
Camoufox is a Firefox fork where fingerprint spoofing happens at the C++ engine level — navigator.*, WebGL, Canvas, AudioContext, screen geometry are all rewritten before any JavaScript can inspect them. That makes it structurally more robust than puppeteer-stealth, which patches the same signals via JavaScript shims that anti-bot systems can detect.
For torch, Camoufox is the answer to "my machine can't run the real-Chrome-clone strategy" — VMs, headless servers, CI pipelines, remote boxes with no host Chrome. It's a strict fallback, not a replacement: when 127.0.0.1:9222 is available (the user's real profile), that's still the better choice because it has real cookies, history, and browsing reputation that Camoufox's synthetic fingerprints cannot replicate.
When to activate
Use this skill when:
- Running on a VM or headless server — no host Chrome to clone, no GUI to drive a headed browser, and
puppeteer.launch()with stealth gets detected on hard sites - Running in CI — same reasons
- You need concurrent personas — torch's cloned profile is a single identity; Camoufox's BrowserForge fingerprint generator produces internally-consistent fresh personas on demand (5% Linux, 2560x1440 at 9.5%, Intel HD GPU at 27.5%, etc. matching real-world distributions)
- The current toolchain is genuinely failing — a site that real-Chrome-clone + stealth + solver + proxy still can't crack (rare, but happens — Interstitial, Kasada, Shape)
Do not use this skill when:
- The user has a real Chrome profile torch can clone (
127.0.0.1:9222is set) — real profile beats synthetic fingerprint for account-level reputation scoring - The site tests specifically for Chrome/Chromium (some sites use V8-specific quirks that SpiderMonkey cannot spoof)
- You only need to scrape soft targets where vanilla stealth plugin already works
Installation
Camoufox is Python-first. torch drives it from Node via the Playwright server mode.
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 · 161 lines · 91 tokens per session scan A 4c49e0e5aa76
camoufox is a skill published in the GitHub repository AgentComputerAI/torch (5 stars, last pushed 4mo ago), licensed MIT. It adds 91 tokens to every session and 2,041 once invoked, about $0.0005 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
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a…
tiny-web-crawler
Crawl from one or more starting web pages, fetch readable content, search within pages, follow relevant links, and stop when the requested information is found or a bounded limit is reached.
tutti-workspace-app-factory
Create, convert, or repair one Tutti workspace app as either a self-contained publishable package under package/ or a Chrome-style local debug app under .tutti/dev-app/. Use for mention://workspace-app-factory/create handoffs, mention://workspace-app-factory handoffs, standalone app generation, adapting existing…
computer-use
Read and drive native desktop applications through the accessibility layer — list on-screen apps, snapshot one window as a numbered element tree, then click / type / set a value / scroll / drag / run a named action, by element index or by screen coordinates. Use for work in a desktop app rather than a web page. Full…
tutti-record-agent-session-replay
From a Tutti checkout, run, audit, freshly replay, publish, or diagnose Session Replay cassettes that are driven by case-repository scenario scripts (CDP), not by interactive UI recording. Use for real-Provider capture while a scenario.mjs executes, cassette transport or semantic-state mismatches, fresh replay…
feature-demo-recording
Record a demo video of a web feature from a real browser. Two modes -- a NARRATED film where measured voiceover drives the timeline (designed slides, subtitles, punch-in camera, rendered from an HTML timeline), and a SILENT evidence clip for a PR or a QA pass. Use when the user asks to record a video, demo, or screen…