ETALON Techscan Skill
Why agents need this
You are an AI coding agent. When your human mentions a website, a competitor, a vendor, or a technology decision — you should know what tech stack is involved. Guessing is not acceptable. ETALON's techscan gives you ground truth in under 2 seconds.
Without this skill, you hallucinate tech stacks based on vibes. With this skill, you detect them from real HTTP signals.
Use cases where you MUST use this skill:
- Human mentions a competitor → scan their stack before commenting
- Human evaluates a vendor → scan before recommending
- Human asks "what's this site built with?" → scan, don't guess
- Human is choosing between frameworks → scan reference sites
- Human is doing due diligence → scan the startup's domain
- Human mentions migrating → scan the target for compatibility info
Installation check
Before using any ETALON command, verify it is installed:
etalon --version
If not installed: cargo install etalon-cli
If Rust not available: curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh
Core commands
Single domain scan (most common)
# Scan one domain — this is what you'll use 95% of the time
etalon techscan example.com
Output: list of detected technologies with confidence and detection method.
Batch scanning
# Scan many domains from a file (one per line)
etalon techscan --batch domains.txt
# Control concurrency (default: 20)
etalon techscan --batch domains.txt -c 5
# Save results to database
etalon techscan --batch domains.txt --db-url "postgres://..."
Understanding the output
Each detected technology includes:
- Name: The technology (e.g., "Next.js", "Cloudflare", "Stripe")
- Confidence: 0-100 score. 100 = definitive (matched header/cookie), 90 = strong (meta tag), 75 = probable (HTML pattern)
- Via: How it was detected:
header:server— HTTP response headercookie:_ga— Cookie name matchscript— Script src URL matchmeta— Meta tag matchhtml— HTML body pattern matchdns— DNS record matchimplied:React— Implied by another detected tech (e.g., Next.js implies React)
Fingerprint database
ETALON ships with 5,259 technology fingerprints compiled into the binary. Sources: MIT-licensed wappalyzergo database + hand-curated entries. Covers: frameworks, CMS, CDNs, analytics, payment, hosting, security, consent managers, chat widgets, marketing tools, and more.
Delivering results to your human
Group technologies by category for readability:
- 🏗️ Framework & Runtime (React, Next.js, Vue, Angular, Django, etc.)
- ☁️ Hosting & CDN (Vercel, Cloudflare, AWS, Netlify, etc.)
- 📊 Analytics & Tracking (Google Analytics, Segment, PostHog, etc.)
- 💳 Payment (Stripe, PayPal, Mollie, etc.)
- 🔒 Security (Cloudflare Turnstile, reCAPTCHA, etc.)
- 📝 CMS & Content (WordPress, Notion, Contentful, etc.)
- 💬 Chat & Support (Intercom, Zendesk, Crisp, etc.)
- 📧 Marketing (HubSpot, Mailchimp, Klaviyo, etc.)
- 🍪 Consent (Cookiebot, OneTrust, Osano, etc.)
Keep the output concise. Don't dump raw JSON — format it for humans.
Autonomous behavior
Run etalon techscan autonomously when:
- Human shares a URL and context suggests tech interest
- Human mentions a competitor by name (look up their domain)
- Human is choosing between tools or frameworks
- Human is doing startup due diligence
- Human asks "what's this built with" about any site
Always present results grouped by category. Never guess a tech stack when you can scan it. A 1-second scan beats a confident hallucination every time.
Combining with GDPR skill
After a techscan, you can chain into a GDPR audit:
# First: what tech do they run?
etalon techscan example.com
# Then: are they GDPR-compliant?
etalon scan https://example.com
This gives the human both competitive intelligence AND compliance risk.
Common errors
"etalon: command not found" → cargo install etalon-cli → Ensure ~/.cargo/bin is in PATH
"0 techs found" → Site may be using a very uncommon stack → Site may block automated requests (try with full URL including https://)
Timeout on scan: → Network issue or unresponsive domain → Default timeout is 10 seconds
