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/wikipedianpx skills add AgentComputerAI/torch --skill wikipediagit 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 | $0.00102 | $0.02091 |
| Opus 5 | $0.00051 | $0.01045 |
| Sonnet 5 | $0.00020 | $0.00418 |
| Haiku 4.5 | $0.00010 | $0.00209 |
Grade A, and why
wikipedia 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)**: Full article HTML returned in ~165 KB, no challenge, no 403. Gate A passed. How it starts
The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Wikipedia (wikipedia.org)
Wikipedia is the easiest possible scrape target: plain server-rendered HTML, generous robots.txt, no bot detection, no rate limiting for reasonable use, and an official API if you want to skip HTML entirely. The only real trap is the 2023+ heading-wrapper div — scrapers that look for
h2as a direct child of.mw-parser-outputwill silently return zero sections.
Detection
| Signal | Value |
|---|---|
| CDN | Wikimedia / Varnish (x-cache, x-varnish) |
| Framework | MediaWiki (server-rendered PHP) |
| Anti-bot | None |
| Auth | Not required for reads |
| robots.txt | Permissive for well-behaved bots; send a descriptive UA |
| JS required | No — full article in initial HTML |
Architecture
Every article lives at https://<lang>.wikipedia.org/wiki/<Title> and is fully server-rendered. The article body is inside:
#mw-content-text > .mw-parser-output
Direct children of .mw-parser-output include <p>, <ul>, <ol>, tables, figures, and — crucially — <div class="mw-heading mw-heading2/3/4"> wrappers around each section heading. Inside that wrapper is the real <h2>/<h3>/<h4> plus an edit-section link. There is no .mw-headline span anymore on modern output; the heading text is just the direct text of the h2/h3/h4.
Reference list: ol.references > li, with the citation text inside .reference-text.
Categories: #mw-normal-catlinks ul li a.
Strategy used
- Phase 0 (curl): Full article HTML returned in ~165 KB, no challenge, no 403. Gate A passed.
- Phase 1: Skipped — no framework JSON to extract; HTML is the source of truth.
- Phase 2: Skipped — no browser needed.
Total time to extract a full article: < 1 second.
Alternative: official APIs (use these for structured data)
If you don't need HTML-level fidelity, skip scraping entirely:
- REST API —
https://en.wikipedia.org/api/rest_v1/page/summary/<Title>returns title, extract, thumbnail, description. - REST API (HTML) —
https://en.wikipedia.org/api/rest_v1/page/html/<Title>returns clean Parsoid HTML (easier to parse than the skinned article page). - Action API —
https://en.wikipedia.org/w/api.php?action=parse&page=<Title>&format=json&prop=sections|text|links|categoriesreturns everything as JSON. - Dumps — for bulk work, use
dumps.wikimedia.orginstead of crawling.
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 · 171 lines · 102 tokens per session scan A 2403bb808aa6
wikipedia is a skill published in the GitHub repository AgentComputerAI/torch (5 stars, last pushed 4mo ago), licensed MIT. It adds 102 tokens to every session and 2,091 once invoked, about $0.0005 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…