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/proxynpx skills add AgentComputerAI/torch --skill proxygit 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 | $0.00083 | $0.02566 |
| Opus 5 | $0.00042 | $0.01283 |
| Sonnet 5 | $0.00017 | $0.00513 |
| Haiku 4.5 | $0.00008 | $0.00257 |
Grade A, and why
proxy 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 yesterday.
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 — 294 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Proxy integration
Proxies are anti-blocking Layer 7 — escalate to them when IP-based blocking is the problem (403/429 on your datacenter IP, rate limits hit fast, geo-blocked content). They don't help with browser fingerprinting or JavaScript challenges — combine with stealth mode (scrape anti-blocking) and CAPTCHA solvers (2captcha, capmonster) as needed.
When to use a proxy
| Symptom | Likely cause | Fix |
|---|---|---|
| 403 on first request | IP reputation (datacenter flagged) | Residential proxy |
| 429 after 5-20 requests | Rate limit per IP | Rotating residential, one IP per request |
| "Not available in your country" | Geo-block | Country-specific exit node |
| Instant Turnstile / JS challenge | Fingerprinting, NOT IP | Don't use proxy — use stealth + solver |
| CAPTCHA appears | Behavior detection | Solver, not proxy (see 2captcha) |
Residential proxies cost money — don't reach for them before trying headed stealth mode and cookie persistence.
Proxy types
| Type | Cost | Trust level | Use case |
|---|---|---|---|
| Datacenter | $0.50-$2/GB | Low (easy to flag) | Low-protection sites, high throughput |
| Residential | $4-$10/GB | High (real ISP IPs) | Protected sites, default choice |
| Mobile (4G/5G) | $10-$20/GB | Highest (shared carrier IPs) | Extreme protection, Instagram, TikTok |
| ISP / Static residential | $1-$3/IP/mo | High, sticky | Authenticated sessions |
Provider comparison (2026)
| Provider | Starting price | Network | Strengths |
|---|---|---|---|
| Smartproxy | $4.50/GB | Large | Best small-scale value, no minimum, simple API |
| Bright Data | $5.88/GB | 150M+ IPs | Largest network, ZIP-code targeting, enterprise compliance |
| Oxylabs | $6.98/GB | 175M+ IPs | Fastest (~0.4s response), 99.82% success rate, $300 minimum |
| IPRoyal | $1.75/GB | Smaller | Cheapest residential, pay-as-you-go |
| Decodo (ex-Smartproxy) | $3.50/GB | Mid | Best value for mid-scale |
Default recommendation for torch: Smartproxy (no minimum, Puppeteer-friendly) or IPRoyal (cheapest). Pick Bright Data only if you need enterprise features or ZIP-precise geo-targeting.
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.
- yesterday First seen · 294 lines · 83 tokens per session scan A 82e7524d0079
proxy is a skill published in the GitHub repository AgentComputerAI/torch (5 stars, last pushed 4mo ago), licensed MIT. It adds 83 tokens to every session and 2,566 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
scroll
Call the unintent tool with the following exact parameters.
mobile-app-ui-design
Design high-quality mobile app UI/UX screens, flows, and components. Use this skill whenever the user asks to design a mobile app screen, create app mockups, build mobile UI components, improve an existing mobile app design, create onboarding flows, design mobile navigation, or requests any mobile-first interface…
proxyclaw
ProxyClaw by IPLoop — Residential proxy access for AI agents. Route requests through 2M+ residential IPs across 195+ countries from real Android, Windows, Mac & Smart TV devices. Bypass geo-restrictions, rotate IPs automatically. Free tier with 0.5GB included. Earn unlimited proxy credits by contributing bandwidth via…
wireless_energy_saving
分析无线节能工单系统的全能助手。包含柱状图统计分析（fetchengblacklistbarchart）和白名单小区列表详情查询（fetchwhitelistcelllist）两大核心功能。该工具通过 CDP 寄生技术直接连接用户浏览器获取鉴权 Token。.
upgrade-browser
Upgrade browser versions (Chrome or Firefox) in the Flutter Web Engine and/or Framework tests. Use when asked to roll or upgrade Chrome or Firefox to a newer version.
generate-token-map
Maps all Figma design tokens from figma-variables.md to their equivalent C++ (Views) identifiers, CSS (WebUI) variables, and Clank (Android) resource attributes/macros/methods in Chromium, and saves the formatted mapping table to the assets directory of the Chrome Design System project. Use this skill when asked to…