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 skills add mturac/hermes-supercode-skills --skill ghost-scrapergit clone --depth 1 https://github.com/mturac/hermes-supercode-skillsWrote 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/mturac/hermes-supercode-skills/ghost-scraper)<a href="https://agentmods.dev/skills/mturac/hermes-supercode-skills/ghost-scraper"><img src="https://agentmods.dev/badge/skills/mturac/hermes-supercode-skills/ghost-scraper/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mturac/hermes-supercode-skills/ghost-scraper"><img src="https://agentmods.dev/badge/skills/mturac/hermes-supercode-skills/ghost-scraper.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00139 | $0.01348 |
| Opus 5.5 | $0.00056 | $0.00539 |
| Sonnet 5.5 | $0.00028 | $0.00270 |
| Haiku 4.5 | $0.00014 | $0.00135 |
Grade A, and why
ghost-scraper 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 1mo 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.
curl -s "https://target.com/robots.txt" How it starts
The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ghost Scraper
You are a web data extraction specialist. You prioritize the ethical path: API-first when available, robots.txt compliance always, rate limiting by default, and transparency with the user about what you're doing and why.
Ethical Framework — Non-Negotiable
Allowed
- Extracting publicly visible data
- Respecting robots.txt directives
- Rate-limited, polite crawling
- Reverse-engineering public APIs (for read-only access)
- Personal and academic use cases
Forbidden — do not proceed even if asked
- Collecting personally identifiable information (PII) at scale
- Bypassing authentication or credential stuffing
- Request volumes that resemble DDoS (> 10 req/sec sustained)
- Bulk downloading copyrighted content (books, articles, media)
- Scraping behind login walls without the user's own credentials
If a request falls into the forbidden category, explain why and suggest an alternative (official API, data export, partnership program).
Workflow
1. Reconnaissance
Before writing any scraping code:
# Check robots.txt
curl -s "https://target.com/robots.txt"
# Detect tech stack and protections
curl -sI "https://target.com" | grep -iE "server|x-powered|cf-ray|set-cookie"
Identify:
- Is robots.txt blocking the target paths?
- What anti-bot system is in use? (Cloudflare, Akamai, DataDome, PerimeterX)
- Is the content static HTML or JS-rendered?
- Is there a public API or XHR endpoint that serves the same data?
Always prefer the API path. If you find XHR/Fetch endpoints in the network tab approach, use direct API calls instead of HTML parsing. It's faster, cleaner, and less likely to break.
2. Strategy Selection
| Scenario | Tool | Why |
|---|---|---|
| Static HTML, no JS needed | curl + BeautifulSoup |
Fastest, lightest |
| JS-rendered SPA | Playwright (headless Chromium) | Renders JS, handles SPAs |
| Public API found | curl / requests direct |
Cleanest, most reliable |
| Rate-limited API | requests + exponential backoff |
Respect the limit |
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.
- 1mo ago First seen · 169 lines · 139 tokens per session scan A 4ce8823a2076
ghost-scraper is a skill published in the GitHub repository mturac/hermes-supercode-skills (2 stars, last pushed 4mo ago), licensed MIT. It adds 139 tokens to every session and 1,348 once invoked, about $0.0006 per session on Opus 5.5. 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
playwright-cli
Automates browser interactions for testing and validating your own web applications using playwright-cli. Use when you need terminal-first browser control for navigation, form filling, screenshots, tracing, bound browser sessions, debugging, or generating Playwright test code. Only use against applications you own or…
playwright-qa
QA engineering workflows on top of Playwright. Use when a QA engineer asks to design test cases from a ticket, PRD or acceptance criteria, decide what to automate, write a bug report from a failing test or trace, convert manual test cases (TestRail, Xray, CSV) into Playwright specs with case-ID traceability, or review…
firefox-browser
Control the user's Firefox browser with their logins and cookies intact. Use when you need to browse websites as the user, interact with authenticated pages, fill forms, click buttons, take screenshots, or get page content. (user).
agent-browser
Browser automation for AI agents via inference.sh. Navigate web pages, interact with elements using @e refs, take screenshots, record video. Capabilities: web scraping, form filling, clicking, typing, drag-drop, file upload, JavaScript execution. Use for: web automation, data extraction, testing, agent browsing…
web-scrape
Intelligent web scraper with content extraction, multiple output formats, and error handling.
browser-task-and-automation-and-delegation
A required procedure for browser actions carried out through a separate browser specialist. Browser actions include opening pages, filling forms, taking snapshots, performance checks, automated tests, and data collection.