Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/chrisdothtml/ai-job-findernpx agentmods add skills/chrisdothtml/ai-job-finder/create-scraperWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/chrisdothtml/ai-job-finder/create-scraper)<a href="https://agentmods.dev/skills/chrisdothtml/ai-job-finder/create-scraper"><img src="https://agentmods.dev/badge/skills/chrisdothtml/ai-job-finder/create-scraper.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00046 | $0.01892 |
| Opus 5 | $0.00023 | $0.00946 |
| Sonnet 5 | $0.00009 | $0.00378 |
| Haiku 4.5 | $0.00005 | $0.00189 |
Grade A, and why
create-scraper 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 8d 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 — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Your goal is to explore a company's careers page and write a permanent, reliable Scraper subclass that other code can use going forward. The argument is either:
- the name of a company, e.g.
/create-scraper netflix - a careers page URL, e.g.
/create-scraper https://explore.jobs.netflix.net/careers, or - a json file containing a list of company names, e.g.
/create-scraper .data/companies.json(in this case, you'll want to go through each company and determine if any existing scrapers work for it, or if a new one needs to be made)
Step 1a - Identify the url for the careers page
If the careers page has been provided as the argument, move to the next step; otherwise search "[company-name] careers" via the WebSearch tool and navigate until you find the careers page.
Step 1b — Identify the company name
Derive the class name and filename from the company's domain or brand name. E.g.:
explore.jobs.netflix.net→NetflixScraper→src/analysis/scraping/NetflixScraper.tsjobs.stripe.com→StripeScraper→src/analysis/scraping/StripeScraper.ts
Step 2 — Navigate and capture network traffic
The goal is to find a JSON API the page calls internally — these are far more reliable than DOM scraping.
- Navigate to the URL with
browser_navigate - Immediately inject a network monitor via
browser_evaluate:
window.__reqs = [];
const _f = window.fetch.bind(window);
window.fetch = function (input, init) {
const url =
typeof input === 'string'
? input
: input instanceof URL
? input.href
: input.url;
window.__reqs.push({ url, method: (init?.method || 'GET').toUpperCase() });
return _f(input, init);
};
const _open = XMLHttpRequest.prototype.open;
XMLHttpRequest.prototype.open = function (method, url) {
window.__reqs.push({ url: String(url), method: method.toUpperCase() });
return _open.apply(this, arguments);
};
- Take a
browser_snapshotto understand the page structure - Interact with the page to trigger data loads: scroll to the bottom, wait a moment, then retrieve what was captured:
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.
- 8d ago First seen · 173 lines · 46 tokens per session scan A d210e5ae1abf
create-scraper is a skill published in the GitHub repository chrisdothtml/ai-job-finder (3 stars, last pushed 1mo ago), licensed MIT. It adds 46 tokens to every session and 1,892 once invoked, about $0.0002 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
browser-workflow-skill-builder
Create robust browser automation skills for sites like LinkedIn, X/Twitter, YouTube, Stripe, or other logged-in web apps by saving page context, analyzing HTML/CSS, writing skill-local JS extract/verify scripts, and using CSS selector actions with screenshot verification.
firecrawl
Search the web and scrape pages into clean markdown with the Firecrawl API — query-based discovery, single-URL extraction including public PDFs, driven by curl with a vault-stored API key.
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…
browser-automation
Playwright-based browser automation patterns for autonomous web interaction.
browser4-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.
hep-browser
Use the Agentlas browser hardpoint for browser-required work.