PUBLICMCP lets any business, library, or training program expose structured data that Claude, GPT, Gemini, and any MCP-compatible agent can query directly. Business identity, services, and location — plus 184,000+ primary sources across peer-reviewed research, historical archives, and a curated semiconductor manufacturing training library for new and aspiring engineers. No scraping. No hallucinating. No dead ends.
AI agents are becoming the new search and learning layer. They research vendors, find service providers, discover nonprofits, surface educational resources, and now teach technical skills to engineers on demand. But today there is no standard way for any organization — business, library, training program, or nonprofit — to make itself discoverable and queryable by these agents.
A PUBLICMCP-compliant server is a lightweight, public, read-only endpoint that exposes structured data in a format any AI agent can query directly. For businesses: identity, services, location, and full site content. For training programs and libraries: primary sources, curriculum sections, and full-text documents — including 187 curated semiconductor manufacturing resources for new and aspiring engineers. No scraping. No authentication. No going stale.
The open standard for the agentic web — so any AI agent can find any organization, access any library, and teach any skill.
PUBLICMCP is the foundation layer. FastMCP gives developers the tools to build compliant servers. Together they form the open infrastructure for how AI agents discover and recommend businesses and nonprofits — and it's forming now.
AI agents get structured, authoritative data directly from your server — not a hallucinated summary scraped from your homepage.
You own your narrative. Your pricing, your services, your service area — updated instantly without touching your website.
AI agents are the new search engines. Businesses queryable today will have a structural advantage as agentic search matures.
Agents shortlisting vendors call get_services on every business in a category and return a structured comparison in seconds.
Any AI agent — Claude, GPT, Gemini, Ollama, or a custom agent — can connect and call these tools:
Business name, description, contact, hours, and publicmcp_version fingerprint.
Full service catalog with pricing ranges and delivery details.
Geographic coverage and service area.
Work examples, filterable by industry.
Markets and verticals served.
Live product catalog. Shopify stores get real-time pricing and inventory — no scraping, no API keys required.
Full-text search across all indexed pages — finds pages by topic, keyword, or location. Ranked to suppress boilerplate noise.
Returns the complete body text of any indexed page — title, meta description, H1, and full content. Used for audits, rewrites, and deep page analysis.
Lists every indexed page with its title and URL. Agents use this to explore site structure before calling get_page_content.
AI-powered project qualification — score a lead against your ideal client profile.
Full-text BM25 search across 184,000+ primary sources — 165k OpenAlex peer-reviewed papers from 50 top universities, 18.5k NYPL digitized historical documents, 187 semiconductor manufacturing training docs, and arXiv preprints. Filter by source collection.
Retrieve a single library item by ID — full record including description, full text (for micron-training items: complete textbook or standard), and attribution URL.
List all 139 curated topic sets in the public library, with item counts and subject tags. Use before search to orient agents toward a specific collection.
Returns the semiconductor training library organized by 7 curriculum sections — Manufacturing Fundamentals, Safety & Compliance, Memory Technology, Industry 4.0, and more. Shows item counts and titles per section. Start here for any semiconductor engineering query.
An agent helping a buyer shortlist vendors can call get_services on every business in a category
and return a structured comparison — no scraping, no hallucination, no dead ends.
Three tool calls to library.publicmcp.org — from graduate-level device physics to OSHA regulatory law to EUV research spanning four universities. Every response cites the primary source. No scraping. No API key.
Engineer: "Pull the MOSFET textbook. Show me the saturation current equation, variable definitions, and a worked 90 nm result."
search_library(query="MOSFET drain current equation", source="micron-training")
library.publicmcp.org/mcp
{ "count": 2, "results": [
{ "item_id": "micron_20260605_194945",
"title": "MOS Transistor",
"description": "Full textbook chapter — I-V characteristics, threshold voltage, body effect.",
"item_type": "book", "institution": "Micron Technology, Inc." }
]}
fetch_library(item_id="micron_20260605_194945")
library.publicmcp.org/mcp
{ "title": "MOS Transistor", "institution": "Micron Technology, Inc.",
"item_type": "book", "source": "micron-training",
"full_text": "...Idsat = W/(2mL) × Coxe × µns × (Vgs − Vt)² [Eq. 6.6.6]
m = 1 + 3Toxe/Wdmax ≈ 1.2 (bulk-charge factor)
[71,015 characters — complete primary source]..." }
| μns | Electron surface mobility (effective mobility) in the inversion layer |
| Coxe | Gate oxide capacitance per unit area (electrical equivalent oxide capacitance) |
| m | Bulk-charge factor: m ≡ 1 + Cdep/Coxe = 1 + 3Toxe/Wdmax ≈ 1.2 |
MOS Transistor — Chenming Hu, Modern Semiconductor Devices for Integrated Circuits, Ch. 6 (2009) · Micron Educator Hub
Engineer: "What's the exact OSHA lockout/tagout sequence before I touch any fab equipment?"
fetch_library(item_id="micron-training_20260607_202917_9c55cb")
library.publicmcp.org/mcp
{ "title": "OSHA 29 CFR 1910.147 — The Control of Hazardous Energy (Lockout/Tagout)",
"institution": "U.S. Department of Labor, OSHA", "item_type": "standard",
"full_text": "...## 1910.147(d) — Application of Control\n
1. Preparation for Shutdown...\n 2. Machine or Equipment Shutdown...\n
3. Machine or Equipment Isolation...\n 4. Lockout or Tagout Device Application...\n
5. Stored Energy...\n 6. Verification of Isolation..." }
OSHA 29 CFR 1910.147 · U.S. Department of Labor, Occupational Safety and Health Administration
Engineer: "Find EUV lithography primary sources across all collections."
search_library(query="EUV extreme ultraviolet lithography semiconductor", source="all")
library.publicmcp.org/mcp
{ "count": 20, "results": [
{ "title": "Extreme ultraviolet lithography reaches 5 nm resolution",
"institution": "Paul Scherrer Institute", "source": "micron-training" },
{ "title": "Maskless extreme ultraviolet lithography",
"institution": "Stanford University", "source": "openalex" },
{ "title": "Time exposure performance of Mo-Au alloys for EUV collector optics",
"institution": "University of Illinois Urbana-Champaign", "source": "openalex" },
{ "title": "Material-specific high-resolution table-top EUV microscopy",
"institution": "Friedrich-Schiller-Universität Jena", "source": "micron-training" },
// ... 16 more
]}
20 results across micron-training and openalex · Call fetch_library on any result for the complete document text
You built a server. Right now it's reachable — but undiscoverable. Claude finds servers
through the registry, not by guessing URLs. A listing means any Claude user who adds
mcp.publicmcp.org/mcp as a connector can find and query your server directly.
Submission takes two minutes. Review and listing within 24 hours.
mcp.publicmcp.org/mcp as a connector, they get two registry tools:
list_registry() returns every listed business — name, hostname, category,
and description. call_business() lets Claude query any of them directly
by name. One connector. Every listing. No URL guessing on their end.
Deploy a PUBLICMCP-compliant server at a public URL with the three required tools: get_info, get_services, get_location.
Fill out the form below with your server URL, business name, and contact email. We verify compliance and add you to the registry within 24 hours.
Any Claude user with mcp.publicmcp.org/mcp connected can call list_registry() to discover your server and query it directly with call_business().
Students using AI to research shouldn't get hallucinated answers because a library's content isn't structured for machine consumption. PUBLICMCP EDU gives public libraries and educational nonprofits a permanently free hosted server — no DNS changes, no technical setup, no cost. Ever.
mcp.[yourname].publicmcp.org| Organization | Type | Endpoint | Tier |
|---|---|---|---|
| PUBLICMCP Public Library 184,000+ primary sources · OpenAlex · NYPL · micron-training · arXiv · 50 Universities |
Digital Library | library.publicmcp.org | open access |
| New York Public Library 1M+ digitized primary sources · 55M catalog items · 92 branches · NYC |
Public Library | nypl.publicmcp.org | edu · free |
| Your library or nonprofit here | — | mcp.[yourname].publicmcp.org | edu · free |
Verified educational nonprofits and public libraries only. One email is all it takes.
The PUBLICMCP Public Library is an open content layer built on the same infrastructure as the standard. Any AI agent can search and retrieve primary sources across six collections: peer-reviewed CS and engineering papers from 50 top universities via OpenAlex (165,876 items), digitized historical records from the New York Public Library (18,541 items), 187 curated semiconductor manufacturing training resources for new and aspiring engineers, and scientific preprints from arXiv — all served via MCP tools, no API key required.
Peer-reviewed CS and engineering papers, digitized historical records from the NYPL, semiconductor training materials, and scientific preprints — across 139 curated topic sets.
Full-depth CS and engineering research from Stanford, MIT, Berkeley, CMU, ETH Zurich, Oxford, Cambridge, Tsinghua, and 42 more top programs worldwide — via OpenAlex (165,876 papers).
187 curated resources for new and aspiring engineers: OSHA safety standards, SEMI industry guides, Micron Technology textbooks, and equipment manuals — organized into 7 curriculum sections via get_training_sections.
BM25-ranked search across titles, abstracts, and full document text. Filter by source collection — use source="micron-training" to search semiconductor content directly.
No API key. No authentication. No rate limits on reads. Any MCP-compatible agent connects to library.publicmcp.org and starts querying immediately.
Every item includes a source_url back to the originating institution — Micron Technology, NYPL, arXiv, OpenAlex, and more. Full primary source, fully attributed.
search_library with a natural language query.
Results include title, description, item type, institution, and source URL — ready for citation.
No paid plan. No API key. Works with any free Claude account — once connected, your Claude has access to 184,000+ primary sources in every conversation.
A free account works. No upgrade required.
Click your profile icon in the top right, then select Settings and choose Connectors from the left menu.
Click Add MCP Server and paste one of the URLs below.
Library only — primary sources & semiconductor training:
https://library.publicmcp.org/mcp
Full stack — library + registry + business identity tools:
https://mcp.publicmcp.org/mcp
After saving, look for the tools indicator near the chat input and toggle your PUBLICMCP server on. This activates the connection for your sessions.
Open a new conversation — your PUBLICMCP tools will be active. Try:
I'm studying semiconductor manufacturing. Show me the training library sections and find beginner content on photolithography.
/mcp.
ChatGPT integrations can reach any PUBLICMCP server via the REST endpoint at /tools/{tool_name}.
You need Python 3.10+, a domain, and a VPS. The entire server is one file.
pip install fastmcp
Fill in your business identity — name, services, location, portfolio. This is the only file you need to change.
{
"name": "Your Business",
"description": "What you do in one sentence.",
"services": [
{ "name": "Service Name", "price_range": "$X–$Y" }
],
"location": { "city": "Boise", "state": "ID" }
}
Point a subdomain at your server, run the server, then submit your listing to appear in the registry.
Inspect how Salt Creative's server is structured as a working reference, or contact Salt Creative for a fully managed deployment.
PUBLICMCP is an open standard for public-facing business and nonprofit identity MCP servers. It defines a required set of tools — get_info, get_services, get_location — that any organization can expose so AI agents can discover and query them without scraping or authentication.
The standard is open — anyone can implement it. Self-hosting a compliant server costs the price of a VPS, typically $5–$20/month. Managed deployments are a paid service offered by Salt Creative.
schema.org embeds structured data in HTML for search engine crawlers. PUBLICMCP exposes live, queryable tools that AI agents call directly via the Model Context Protocol — no scraping required. Responses can include real-time data like current pricing and availability, and update instantly without touching your website.
Any agent that supports the Model Context Protocol: Claude, GPT, Gemini, Ollama, and custom agents. If the agent can connect to an MCP server, it can query any PUBLICMCP-compliant server natively via /mcp.
For browser chat users: Claude.ai is the only mainstream browser UI with native MCP connector support. Developers using the Gemini API, Ollama, or any MCP-compatible framework connect via /mcp. ChatGPT integrations can use the REST endpoint at /tools/{tool_name}.
Open Claude.ai → click your profile icon → Settings → Connectors → Add MCP Server → paste https://library.publicmcp.org/mcp → Save. That's it. No paid plan required — a free Claude account works. Your Claude will have access to 184,000+ primary sources — including the semiconductor manufacturing training library — in every conversation from that point forward.
For self-hosting, you need Python 3.10+, a VPS, and a domain. FastMCP is the reference runtime — install it with pip install fastmcp, configure your business.json, and deploy. If that's out of reach, Salt Creative offers fully managed deployments typically live within 48 hours.
Required tools (get_info, get_services, get_location) must be implemented for spec compliance. Recommended tools (get_portfolio, get_markets, get_products, search_site_content, get_page_content, list_site_pages) are strongly encouraged and included in the reference implementation but are not mandatory for a compliant server.
Every registry entry requires a live compliant server at a public URL. If you can't self-host, Salt Creative's managed hosting service covers setup, hosting, and maintenance so your business appears in the registry without running any infrastructure yourself.
PUBLICMCP servers are intentionally public and read-only — they expose only what you choose to include in your business.json. No authentication is required for agents to query your server, and no customer data or private information should ever be included. Think of it as a public business profile, not a database.
Yes. PUBLICMCP is an open standard — anyone can implement a compliant server on any stack. The standard is designed to evolve alongside the Model Context Protocol, which is already incorporated into official Anthropic, OpenAI, and Google AI SDKs. FastMCP, the leading MCP framework with 1M+ daily downloads, serves as the reference runtime. The foundation is solid.
Salt Creative offers fully managed PUBLICMCP deployments — setup, hosting, and maintenance included.
Learn about managed hosting →