Universal Native MCP Server Hub
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| topai_web_to_markdownA | Converts any public HTTP/HTTPS webpage into clean, ad-free, token-optimized Markdown with document metadata. Use when an agent needs to read article text, documentation, or web pages without HTML/DOM clutter before RAG indexing or summarization. Do not use for capturing visual pixel screenshots (use |
| topai_b2b_data_leadsA | Queries verified B2B company and executive lead records filtered by industry vertical, ISO country code, and result limit. Use when building prospect lists, sales pipelines, or market landscapes across specific sectors. Do not use when you already have a specific company domain and want deep technographic profiling on that single domain (use |
| topai_company_enrichmentA | Enriches a single corporate domain name with firmographic metadata including legal entity name, industry sector, workforce size range, detected software stack, and social links. Use when you have a known company domain (such as 'linear.app') and need firmographic context. Do not use for discovering lists of unknown companies by industry (use |
| topai_ai_web_searchA | Executes a real-time web search query and returns noise-free organic results with page titles, canonical URLs, and factual snippets for LLM grounding. Use when an agent needs up-to-date external facts, news, or URL discovery across the open web. Do not use when you already have the exact URL to read (use |
| topai_semantic_cacheA | Performs sub-5ms semantic vector cache lookups ('get') and completions storage ('set') for LLM prompts to eliminate redundant inference latency and token spend. Use before invoking an expensive upstream LLM to check if an equivalent prompt was already answered, or right after generation to cache the completion. Do not use for compressing raw text length within a single uncached prompt (use |
| topai_invoice_extractorA | Parses unstructured invoice or receipt text into structured accounting fields including vendor name, invoice number, ISO date, currency, VAT amount, gross total, and line items. Use when extracting structured financial data from OCR output or raw invoice text. Do not use for verifying mathematical VAT calculations or reconciling line-item arithmetic (use |
| topai_headless_screenshotA | Renders a public webpage in a serverless headless browser and returns viewport or full-page screenshot metadata and dimensions. Use when visual layout inspection, UI verification, or image capture of a live URL is required. Do not use when you only need the textual content of a webpage (use |
| topai_prompt_shieldA | Scans user prompts or untrusted external text in <1ms to detect adversarial prompt injections, jailbreak attempts, toxic payloads, and PII leakage. Use as a pre-inference security guardrail before passing untrusted user input or scraped web content to an LLM. Do not use for compressing safe prompts (use |
| topai_token_slimmerA | Compresses bloated HTML, JSON, or plain-text payloads losslessly to reduce LLM input token consumption by 30% to 70% without losing semantic meaning. Use before sending large JSON API responses, logs, or RAG chunks into an LLM context window. Do not use for converting a live URL directly into Markdown (use |
| topai_agentvision_cropA | Computes optimal Region-of-Interest (ROI) bounding boxes for UI buttons, tables, charts, or text blocks in an image to reduce multimodal vision token consumption by up to 75%. Use when preparing screenshots or document images for vision-capable LLMs so only the relevant region is sent. Do not use for taking the initial webpage screenshot (use |
| topai_math_fact_checkerA | Executes deterministic IEEE/decimal financial arithmetic for VAT calculation, multi-line invoice reconciliation, and working business-day counts with an explicit step-by-step audit trail. Use whenever an agent must compute or verify taxes, invoice totals, or date intervals without risking LLM arithmetic hallucinations. Do not use for parsing raw unstructured invoice text (use |
| topai_techstack_fingerprintA | Fingerprints a web domain in <50ms across 70+ technology signatures to identify its CMS, e-commerce engine, analytics stack, CDN, and frontend frameworks. Use when qualifying technical architecture or detecting whether a site runs Shopify, Stripe, HubSpot, Next.js, or Cloudflare. Do not use when you need general workforce size or executive firmographics (use |
| topai_polymarket_arbitrageA | Computes prediction market implied probability, mispricing delta, expected value (EV%), and fractional Kelly criterion position sizing between a Polymarket contract price and a reference benchmark probability. Use for quantitative prediction market analysis, odds comparison, and risk-adjusted bankroll allocation. Do not use for standard VAT or invoice accounting (use |
| topai_mcp_hub_registryA | Introspects and searches the TopAI Sovereign Gateway catalog of MCP tools by functional category or keyword. Use when an autonomous agent wants to discover which specialized tools are available on this hub before selecting one. Do not use to execute a domain tool directly—call the specific |
| topai_pharmapatent_watcherA | Retrieves pharmaceutical patent cliff dates, active ingredient metadata, FDA Orange Book / European exclusivity expirations, and generic entry threat levels for a given drug name. Use for life-sciences competitive intelligence, LOE (Loss of Exclusivity) forecasting, and pharmaceutical IP analysis. Do not use for verifying individual physician NPI medical licenses (use |
| topai_healthcare_validatorA | Validates 10-digit US National Provider Identifiers (NPI) using the ISO Luhn 80840 prefix checksum algorithm and returns provider license status and specialty metadata. Use when verifying healthcare practitioners, medical facilities, or compliance onboarding records. Do not use for looking up drug patent expiration dates (use |
| topai_google_ai_overviewA | Extracts Google Search Generative Experience (AI Overview) synthesis text, explicitly cited source URLs, and organic SERP count for a search query. Use for GEO (Generative Engine Optimization) audits, citation tracking, and analyzing how Google AI summarizes a topic. Do not use for general fast web search when AI Overview citation tracking is unnecessary (use |
| topai_disposable_email_shieldA | Detects disposable, burner, or temporary email addresses using RFC 5322 syntax parsing, throwaway domain blocklists, and DNS MX heuristics, returning a fraud risk score (0.0 to 1.0). Use during user signup, B2B lead cleaning, or KYC fraud prevention to block fake accounts. Do not use for enriching corporate firmographics when you already know the domain is legitimate (use |
| topai_openapi_flattenerA | Dereferences internal and external |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 19 tools
Every tool has a clearly distinct purpose, reinforced by explicit 'Do not use ... instead' cross-references in descriptions. Overlaps are minimal and well-handled (e.g., search vs markdown conversion vs AI Overview extraction).
All tools follow a consistent topai_ prefix and snake_case naming pattern. The convention is predictable and uniform across the set.
19 tools is slightly above the ideal 3-15 range, but appropriate for a 'universal hub' meant to aggregate many specialized utilities. Each tool earns its place; no obvious filler.
The hub covers a broad range of stateless read-only utilities (web, search, security, data enrichment, text compression, math, vision). While no single domain has full CRUD, the set is coherent for a utility hub and includes a registry tool for discovering more.