Micropayment API network powered by x402 protocol. Exposes 16 utility and AI tools as MCP tools for AI agents. Uses stdio transport for Claude Desktop / ChatGPT MCP integration.
16 tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. All tools have descriptions (10-200 chars, mostly adequate), and input schemas are present with Zod validation. However, most parameter descriptions are minimal or missing context about expected formats, ranges, and constraints. No documented output schemas. Error handling is generic (JSON stringification). Tool naming is mostly clear (verb_noun pattern), but some tools like 'scrape' and 'html_to_markdown' overlap in purpose. No tool annotations (readOnlyHint, destructiveHint). Composition is reasonable, tools are single-purpose, but chaining support is unclear due to undocumented output fields.
Generate AI-powered content using Llama 3.3 70B. Supports blog posts, product descriptions, email copy, social media posts, landing page copy, and SEO articles.
Encode text to base64, or decode a base64 string back to text.
Look up DNS records (A, AAAA, MX, TXT, NS) for a domain. Returns all available record types.
Comprehensive domain email health analysis: checks MX records, SPF, DKIM, DMARC, and DNS resolution. Returns a score (0-100) with actionable improvement tips.
Fetch a web page and convert its HTML content to clean Markdown text. Useful for reading web pages in a text-friendly format.
Fetch and return the HTTP response headers for any URL. Useful for debugging CORS, caching, security headers, and content types.
No documented output schemas. Tools return raw JSON from backend API without specifying field names, types, or structure. LLMs cannot plan downstream calls or extract specific fields reliably.
Parameter descriptions lack format/constraint details. E.g., 'domain' params do not specify format (FQDN, no protocol), 'url' params lack protocol requirement, 'size' in qr_code has no min/max bounds. LLMs cannot validate inputs before calling.
Overlapping tool purposes without clear disambiguation. 'scrape' and 'html_to_markdown' both fetch and process web pages; 'url_meta' and 'scrape' both extract page data. Descriptions do not explain when to use each.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 56 | 2026-07-28+ | v2 |
Get information about an IP address: reverse DNS hostnames and IPv4/IPv6 type detection.
Format, validate, and optionally minify JSON data. Accepts a JSON string or object and returns prettified (or minified) output with byte size.
Analyze the strength of a password. Returns a score (0-8), strength level, length, and improvement suggestions.
Generate a QR code as a base64-encoded PNG image. Accepts any text or URL.
Scrape a web page and extract structured data: clean text content, page title, meta description, headings, links, and images.
Check SSL/TLS certificate details for a domain: validity, issuer, subject, expiry dates, serial number, and fingerprint.
Analyze text and return statistics: character count, word count, sentence count, paragraph count, estimated reading time, and speaking time.
Extract metadata from a URL: page title, meta description, Open Graph image, OG title, OG type, and favicon URL.
Verify an email address: syntax validation, MX record check, disposable email detection, role-based address check, and free provider detection. Returns a score from 0-100.
Perform a WHOIS lookup for a domain. Returns registrar, creation date, expiry, nameservers, and other registration details.
Generic error handling. All errors return raw JSON or error message text without recovery guidance. No categorization (retryable vs user-fixable vs fatal). LLMs cannot determine next action.
No tool annotations. Tools lack readOnlyHint, destructiveHint, or idempotentHint metadata. LLMs cannot infer safety properties or retry behavior.