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/githubnpx skills add AgentComputerAI/torch --skill githubgit 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 | $0.00063 | $0.01151 |
| Opus 5 | $0.00032 | $0.00575 |
| Sonnet 5 | $0.00013 | $0.00230 |
| Haiku 4.5 | $0.00006 | $0.00115 |
Grade A, and why
github 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Phase 0 curl → done. Data is in raw HTML, no protection headers, no JS required. Skip browser entirely. How it starts
The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GitHub (github.com)
GitHub's public HTML pages are fully server-rendered with no JS gating and no bot protection on unauthenticated GETs. A one-shot
fetch+ cheerio parse is all you need. Only reach for the REST API (api.github.com) when you hit the 60 req/hr unauthenticated limit or need structured fields (topics, license, default branch) not present on the HTML.
Detection
| Signal | Value |
|---|---|
| CDN | GitHub edge (Varnish/Fastly) |
| Framework | Rails + Turbo, React islands |
| Anti-bot | None on public pages |
| Auth | Optional — anon works |
| Rate limit (HTML) | Generous, IP-based, soft |
| Rate limit (REST) | 60/hr unauth, 5000/hr with token |
Architecture
/trendingis a normal Rails view. Repo cards are<article class="Box-row">inside<div data-hpc>.- Note: the trending list size varies — historically 25, but currently GitHub often returns only ~10 cards. Don't hardcode the count.
- Repo pages (
/{owner}/{name}) have data in multiple places: HTML meta tags,<react-app>embedded JSON payloads, and DOM nodes. For simple fields (stars, forks, description), raw DOM is easiest. - Sitemaps at
/sitemap.xmlare huge — prefer targeted URL lists.
Strategy used
Phase 0 curl → done. Data is in raw HTML, no protection headers, no JS required. Skip browser entirely.
Extraction (trending)
import { load } from 'cheerio';
const res = await fetch('https://github.com/trending', {
headers: { 'User-Agent': 'Mozilla/5.0' },
});
const $ = load(await res.text());
$('article.Box-row').each((_, el) => {
const $el = $(el);
const fullName = $el.find('h2 a').attr('href').replace(/^\//, '').replace(/\s+/g, '');
const description = $el.find('p').text().trim() || null;
const language = $el.find('[itemprop="programmingLanguage"]').text().trim() || null;
const stars = parseInt($el.find('a[href$="/stargazers"]').text().trim().replace(/,/g, '')) || 0;
const forks = parseInt($el.find('a[href$="/forks"]').text().trim().replace(/,/g, '')) || 0;
const starsToday = $el.find('span.d-inline-block.float-sm-right').text().trim().replace(/\s+/g, ' ');
// ...
});
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 · 105 lines · 63 tokens per session scan A f6796d9f6be2
github is a skill published in the GitHub repository AgentComputerAI/torch (5 stars, last pushed 4mo ago), licensed MIT. It adds 63 tokens to every session and 1,151 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other skills, from other repositories
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.
playwright-cli
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
browserstack
../../../engineering-team/playwright-pro/skills/browserstack/SKILL.md.
analyze-performance-traces
Analyze Chrome, Chromium, Electron, React DevTools, or Perfetto-compatible JSON traces and audit user-reported profiling findings without loading large artifacts into context; prove trigger-to-render/layout chains, separate measured facts from source inference, find exact code choke points, classify forced layout and…
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…
browser-recording
Record a browser flow as a video/GIF for evidence — animations, transitions, and multi-step interactions that a still screenshot cannot prove. Drives the project's own Playwright through a bundled runner, then converts to mp4 + GIF via ffmpeg. Use when the user asks to record a demo, capture a GIF or video of the UI…