Website Intelligence + Marketing API Bundle exposed as MCP tools. Provides 17 APIs across 2 droplets for website analysis, contact extraction, technology detection, DNS/SSL checks, and marketing intelligence.
Server exposes 12 web intelligence and marketing tools with complete input schemas and descriptions. All tools are READ_ONLY and follow verb_noun naming (website_to_markdown, contact_extractor, etc.). Descriptions are substantive (100-200 chars) and explain WHAT the tool does and WHEN to use it. However, output schemas are NOT documented, LLMs cannot predict response structure or plan downstream calls. Parameters lack type constraints (enums, ranges, patterns). Error handling is absent from visible code. No tool annotations (readOnlyHint, idempotentHint). Security: API keys are injected server-side (good), but no audit logging visible. Composition is sound, each tool has one responsibility and tools chain naturally (e.g., contact_extractor → marketing_contact_extractor). Baseline: 549 production tools average 18-char names and 194-char descriptions; this server meets those targets but lacks output documentation and error guidance.
Generate a structured AI-powered summary of any website: company overview, industry classification, key offerings, target audience. Optionally generate an AI narrative with business analysis. Costs more due to LLM inference.
Extract all contact information from a business website: email addresses, phone numbers, social media links (LinkedIn, Twitter, Facebook, Instagram, YouTube, GitHub), and physical addresses. Supports deep crawling of /contact, /about, /team pages.
Perform comprehensive DNS lookups for any domain: A, AAAA, MX (mail servers), NS (name servers), CNAME, TXT records. Also discovers common subdomains (www, mail, blog, api, etc.) and checks DNSSEC status.
[Marketing Bundle] Extract reviews, ratings, and business metadata from Google Maps listings. Search by Place ID or by business name + location. Returns ratings, review counts, top reviews, hours, address, phone, and website.
[Marketing Bundle] Extract business contact information from any website: email addresses, phone numbers, social media profiles (LinkedIn, Twitter, Facebook, Instagram), and physical address.
Output schemas not documented. LLMs cannot predict response structure, field names, or types. Prevents planning downstream tool calls and forces trial-and-error parsing.
No error handling guidance in tool definitions. Errors from external APIs (timeouts, 404s, rate limits) are not documented. LLMs cannot distinguish retryable from fatal failures.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 65 | 2026-07-28+ | v2 |
Extract Open Graph (og:) and Twitter Card (twitter:) tags from any website. Returns social preview data: title, description, image, URL, site_name, type, and Twitter-specific tags.
Fetch and parse robots.txt from any website. Returns crawl rules per user-agent, disallowed/allowed paths, crawl delays, sitemap URLs, and accessibility analysis for key pages.
Discover and parse XML sitemaps from any website. Supports standard sitemap.xml, sitemap index files, and common alternative sitemap paths. Returns all URLs with lastmod dates, change frequency, and priority.
Check SSL/TLS certificate details for any domain. Returns: issuer, subject, validity period, days to expiry, cipher suite, protocol (TLS 1.2/1.3), certificate type, security grade (A+ through F), and SANs.
Detect the technology stack used by any website: CMS, JavaScript frameworks, analytics tools, CDN providers, web servers, and UI libraries. Returns confidence scores per technology.
Extract comprehensive metadata from any website: meta tags, OpenGraph, Twitter Cards, headings (H1-H6), images (with counts/SRSet), canonical URL, language, favicon, and structured data.
Convert any web page to clean, structured Markdown. Returns the page title, description, and full markdown content.
No tool annotations (readOnlyHint, idempotentHint, destructiveHint). All tools are READ_ONLY but this is not declared in the tool definition. Agents cannot infer safety properties.
Parameter constraints missing. 'url' and 'domain' parameters accept free-form strings with no format validation, length limits, or pattern constraints. LLMs may pass invalid URLs.
google_maps_reviews has ambiguous parameter logic. 'place_id' and 'business_name' are both optional but interdependent. No description clarifies which is required or how the tool behaves if both/neither are provided.