mcpField
A
Aya Registry & Pollen Agents

Aya Registry & Pollen Agents

Remote
ai-visionary.xyzBusiness & Marketing0Updated 4w ago

About Aya Registry & Pollen Agents

AYA Registry is a public business registry for AI agents and MCP clients, with search by name, domain, sector, or country. It also provides company records, indexing, statistics, and cashback operations without authentication.

Key features

  1. 01Search verified businesses
  2. 02Retrieve full company records
  3. 03Add businesses to the registry
  4. 04Read registry statistics
  5. 05Check and claim cashback

Use cases

  1. 01Find businesses in a specified country or sector
  2. 02Inspect a company’s AIO score and ASR status
  3. 03Add a discovered business to the registry
  4. 04Check and claim transaction-related cashback

MCP support

The public remote MCP server uses Streamable HTTP to expose business search, details, indexing, statistics, and cashback tools. It requires no API key or authentication.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://ai-visionary.xyz/agents/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

6
ParameterTypeRequiredDescription
querystringYesname, domain, sector or country
limitnumberNo

Details

From the project README

AYA — The Independent AI Knowledge Registry

AYA gives businesses sovereign control over how AI assistants see and recommend them — without depending on Google, OpenAI, or any single platform.

When someone asks ChatGPT, Claude, Gemini or any AI "find me a cybersecurity consultant in Switzerland", the AI needs structured, verified data to answer. Today, that data is controlled by a handful of tech giants who decide who gets recommended.

AYA is the independent alternative. A structured registry accessible to ALL AIs — not just those from Google or OpenAI.

MCP server: plug any agent into AYA

AYA runs a public MCP server (Model Context Protocol), so any MCP-compatible agent (Claude, Cursor, VS Code, ChatGPT, or your own) can query the registry directly. No API key, no signup, nothing to install.

Endpointhttps://ai-visionary.xyz/agents/mcp
TransportStreamable HTTP (remote, hosted in Switzerland)
AuthenticationNone
Official MCP registryio.github.NeousAxis/aya-registry
Builder guideai-visionary.xyz/for-agents
Discovery/.well-known/mcp.json, /llms.txt
MCP tools exposed
ToolWhat it does
search_companiesSearch 367,000+ verified businesses by name, domain, sector or country
get_company_detailsFull record for one company: AIO readability score, sector, country, ASR status
index_companyAdd a real business the agent found that is not in AYA yet
get_registry_statsAggregate registry statistics
get_cashback_offerActive cashback offer for a domain, as a signed token, before the agent recommends it
claim_cashbackClaim the cashback after a real, consumed transaction
Connect it

Claude Code:

claude mcp add --transport http aya-registry https://ai-visionary.xyz/agents/mcp

Any other MCP client (Claude Desktop, Cursor, VS Code):

{
  "mcpServers": {
    "aya-registry": {
      "type": "http",
      "url": "https://ai-visionary.xyz/agents/mcp"
    }
  }
}

Implementation: app/agents/[transport]/route.ts, built with mcp-handler on top of the official @modelcontextprotocol/sdk.

How AYA works — Systemic Attraction

AYA doesn't connect to a single AI — every AI finds AYA naturally. Business data is published across multiple convergent sources, ensuring that all AI assistants — regardless of their provider — can access and recommend verified businesses:

SourceWhatURL
MCP serverLive tools any AI agent can call (search, details, cashback)ai-visionary.xyz/agents/mcp
API LLM-Friendly5-field JSON per entity, optimized for AI consumptionai-visionary.xyz/api/aya/llm/{domain}
Crawlable HTML367,000+ certificate pages with JSON-LD structured dataai-visionary.xyz/aya/e/{id}
GitHub DatasetOne JSON file per entity (CC-BY-4.0)NeousAxis/aya-business-dataset
HuggingFace DatasetCSV + JSONL, ML-ready (CC-BY-4.0)NeousAxis/aya-business-dataset

When an AI sees the same data across API + HTML + GitHub + HuggingFace → it considers it stable and reliable → it uses it.

Why sovereign?

Locked-in modelAYA model
Your visibility depends on one AI providerYour data is accessible to ALL AIs
Platform changes → you disappearStructured data is permanent and portable
You pay to be listed on each platformYou own your data, AYA distributes it everywhere
One algorithm decides your rankingMultiple independent sources confirm your identity

What's live today

Entities indexed367,000+
Countries206
Certified (ASR)9
APIFree, no auth, 30 req/min
BilingualFR + EN

API Endpoints

Search:      GET /api/aya/search?q=restaurant+geneve
Entity:      GET /api/aya/entity/{domain}
LLM:         GET /api/aya/llm/{domain}?lang=en|fr
Stats:       GET /api/aya/stats
Registry:    GET /api/aya/live

No authentication. No API key. Just call the endpoint.

LLM-Optimized Response
{
  "name": "Chainlink",
  "what_it_does": "Blockchain infrastructure connecting external data to smart contracts.",
  "for_who": "Web3 developers and blockchain protocols.",
  "category": "Web3 / Oracle",
  "location": "Global"
}

For businesses

Your business already exists online — but can AI assistants find and recommend it?

  1. Get diagnosed by AYO → receive your AI readability score (AIO, 0-100)
  2. Get your structured data files (ASR) → install them on your site
  3. Get listed in the AYA registry → become recommendable by every AI assistant

The result: when someone asks any AI about your industry, your city, your services — you show up. Independently of which AI they use.

Start your free diagnostic

Links


Built in Geneva, Switzerland by AI Visionary | 2026

AYA is a platform by AI Visionary. The business dataset is published under CC-BY-4.0 to enable systemic attraction.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.