domaininfo#

Look up domain WHOIS information and (optionally) capture website screenshots.

When to Use#

When the user types whois <domain> or whois <url> or whois <email>:

• Strip any https://, http://, www. prefixes
• If input contains @, extract the domain part after @ (e.g., user@example.com → example.com)

Security Considerations#

• Input validation: After extracting the domain, only allow alphanumeric, hyphen, and dot characters. Reject anything else.
• Command injection prevention: Never interpolate user input directly into shell strings. Prefer argument arrays / safe libraries.
• Timeouts: Every external call must have a bounded timeout (e.g., 10s for WHOIS/DNS, 10s for TLS).
• Error handling: On failure, return a generic user-friendly message and keep details internal.
• Output sanitization: Build the final message as a single string before sending; never send partial responses.
• File-system safety: If writing screenshots, restrict writes to a known directory under the skill folder and verify paths stay within it.
• Rate-limiting & caching: Cache IP-to-country lookups briefly to avoid hammering external services.

Workflow (Strict Buffer-First — SAFE EXECUTION)#

CRITICAL: Zero output until everything is ready. No progress messages.

Phase 1 — Silent Buffer with Validation#

1. Extract & validate domain
   • Strip https://, http://, www. prefixes.
   • If input contains @, take the part after @.
   • Validate with regex ^[a-z0-9.-]+$ (case-insensitive).
   • If invalid, abort and return “❌ Invalid domain”.
2. WHOIS: run whois via safe exec with timeout (10s). Store registrar data.
3. DNS: run dig for A, AAAA, NS, MX via safe exec with timeout (10s). Store results.
4. IP Geolocation (Country Code)
   • For each IP from A/AAAA and resolved NS/MX hostnames:
     • Query https://ipinfo.io/{IP}/country using web_fetch with timeout (5s).
     • Store the returned 2-letter country code.
5. Email Security (DMARC/SPF/DKIM)
   • DMARC: query TXT for _dmarc.<domain>
   • SPF: query TXT for <domain> and extract the string containing v=spf1 (parse in code; avoid shell pipelines)
   • DKIM: query TXT for common selectors ( default, google, selector1)

Phase 2 — Optional Screenshot + TLS#

Screenshot (ONLY if screenshot tooling is already available)

Only attempt a website screenshot if one of the following is already available in this runtime:

• OpenClaw browser tool (preferred): use the browser tool to navigate to the site and take a screenshot.
• Bundled Playwright script: scripts/domain-screenshot.js (only if Node + Playwright + a Chromium runtime are already installed).

If neither is available (missing tool / missing module / missing browser runtime), skip the screenshot silently and continue the report.

TLS/SSL Check (if HTTPS)

• Fetch certificate info with openssl (timeout 10s).
• Extract: certificate issuer and expiry date.
• If it fails or times out, note “TLS check failed” but continue.

Phase 3 — Single Final Output#

• If a screenshot was successfully captured, send it via the message tool.
• Send the final WHOIS + DNS + Email Security + TLS summary in one message only.

Send Screenshot (SINGLE SEND ONLY)#

Use message tool with action=send and filePath:

json

{
  "action": "send",
  "caption": "domain.com screenshot",
  "filePath": "domain-screenshot.png"
}

Do NOT also implement provider-API fallbacks (e.g., raw HTTP requests). If message sending fails, report failure rather than double-sending.

Setup Notes#

• This skill does not include step-by-step installation instructions for Playwright/Chromium.
• Screenshot is an optional enhancement and must be skipped if screenshot tooling is not already present.
• See references/setup.md for non-invasive environment notes.