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/yelpnpx skills add AgentComputerAI/torch --skill yelpgit 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 | $0.00125 | $0.02391 |
| Opus 5 | $0.00063 | $0.01196 |
| Sonnet 5 | $0.00025 | $0.00478 |
| Haiku 4.5 | $0.00013 | $0.00239 |
Grade A, and why
yelp 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
description: Proven scraping playbook for yelp.com search result pages (/search?find_desc=...&find_loc=...). Yelp is behind DataDome — bare curl gets HTTP 403 with `server: DataDome` and a captcha-delivery interstitial. How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Yelp (yelp.com)
Yelp's SERP is fully server-rendered HTML wrapped in a DataDome anti-bot. The trick is not to fight DataDome — connect to the user's real Chrome and the cookie does the work.
Detection
| Signal | Value |
|---|---|
| CDN | Fastly (x-served-by: cache-pao-...) |
| Framework | SSR HTML (no __NEXT_DATA__, no React shell) |
| Anti-bot | DataDome (server: DataDome, x-datadome: protected, x-dd-b: 2) |
| Auth | Not required for search |
| Structured | ld+json BreadcrumbList + FAQPage only — no ItemList |
| robots.txt | Disallows /search for most bots, but real users are fine |
curl -sL https://www.yelp.com/search?find_desc=pizza&find_loc=New+York%2C+NY → HTTP 403 with a geo.captcha-delivery.com iframe in the body.
Architecture
- Pure SSR. The full result list ships in the initial HTML — no XHR hydration.
- Each page returns 10 organic results plus ~13 ad cards (
[data-testid="serp-ia-card"]). - Pagination is
&start=10,&start=20, … (offset, not page number). - DataDome enforcement is per-session: the first navigation in a fresh browser context triggers a JS challenge, but a real Chrome with a persistent profile already has a valid
datadomecookie and skates through. Disposable Chromium with stealth is NOT enough — DataDome scores it as a bot.
Strategy used
- Phase 0 (curl): 403 DataDome → escalate.
- Phase 1 (framework): no Next/Nuxt blob, no API endpoint visible in source → escalate.
- Phase 2 (browser):
puppeteer.connect({ browserURL: "http://127.0.0.1:9222" })→ first nav clears DataDome automatically within ~1.5s, subsequent navs are instant. - Phase 3: parse with cheerio, walk leaf-text spans for neighborhood/status/snippet.
Stealth config that works
None. Don't use puppeteer-extra/stealth on the real-Chrome path — Yelp's profile cookie is the entire bypass. Just:
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 · 163 lines · 0 tokens per session scan A 493cad2260be
yelp is a skill published in the GitHub repository AgentComputerAI/torch (5 stars, last pushed 4mo ago), licensed MIT. It adds 125 tokens to every session and 2,391 once invoked, about $0.0006 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.
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…
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…
browserstack
../../../engineering-team/playwright-pro/skills/browserstack/SKILL.md.
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…