acsearch-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@acsearch-mcpsearch for 'Athena owl tetradrachm'"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
acsearch-mcp
MCP server for acsearch.info — the world's largest ancient coin auction archive, with over 16 million lots from 475 auction houses spanning 20+ years.
Give your AI agent the ability to search auction records, retrieve lot details with specs and references, analyze price history, and find comparable sales — all through acsearch.info's Premium database.
Framework-agnostic: works with any MCP-compatible client including Hermes Agent, Claude Desktop, Claude Code, Cursor, and Continue.dev.
Quick Start
uvx acsearch-mcp --transport stdioRequires an acsearch.info Premium account (€85/year) for price data. Set your credentials:
export ACSEARCH_USERNAME="your_username"
export ACSEARCH_PASSWORD="***..."Cookies are cached to ~/.acsearch_cookies.json after first login — no repeated authentication on every tool call.
Related MCP server: mcp-server-wayback
Tools
Tool | Description |
| Full-text Boolean search across 16M+ auction lots |
| Full lot details: description, specs, references, provenance |
| Year-by-year price analysis for market research and valuation |
| Recent comparable auction results sorted by date |
Search Syntax
acsearch.info uses Boolean full-text search:
Operator | Example | Effect |
|
| Matches Philip, Philippe, Philippus |
|
| Exact phrase match |
|
| Exclude results containing the word |
|
| Match either term |
Examples: "Philip II tetradrachm", Alexander* stater -bronze, cistophorus Ephes*
Client Configuration
Hermes Agent
mcp_servers:
acsearch:
command: uvx
args:
- acsearch-mcp
- --transport
- stdio
env:
ACSEARCH_USERNAME: your_username
ACSEARCH_PASSWORD: ***Restart: hermes gateway restart or /reset in CLI.
Claude Desktop
{
"mcpServers": {
"acsearch": {
"command": "uvx",
"args": ["acsearch-mcp", "--transport", "stdio"],
"env": {
"ACSEARCH_USERNAME": "your_username",
"ACSEARCH_PASSWORD": "your_password"
}
}
}
}Any MCP Client (Generic)
uvx acsearch-mcp --transport stdioWith ACSEARCH_USERNAME and ACSEARCH_PASSWORD set in the environment.
HTTP Mode
For remote or networked clients:
uvx acsearch-mcp --transport http --host 0.0.0.0 --port 8001Connect to http://host:8001/mcp.
From Source
git clone https://github.com/wushanyun64/acsearch-mcp.git
cd acsearch-mcp
uv sync
uv run acsearch-mcp --transport stdioHow It Works
acsearch.info embeds all search results as structured JSON in a <script> tag:
acsearch.initSearchResults = [
{ "id": "2581457", "title": "Heritage Auctions, Auction 3041, Lot 32015",
"description": "MACEDONIAN KINGDOM. Philip II...",
"date": "13.08.2015", "price": "6'250 USD", "image": "https://..." }
];The parser extracts this JSON directly — no fragile HTML scraping needed. The client authenticates with your Premium credentials, maintains a cookie-backed session, and enforces polite rate limiting (2 seconds between requests).
Limitations
Descriptions are capped at 350 characters by acsearch.info. The full catalog text is available on the original auction house page (linked in the
linkfield of lot details).Prices require a Premium account.
License
MIT © 2026 Jason Sun
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP server for Open Archives: Dutch genealogical records and historical page transcriptions.
- Prowl MCPOAuthchat.prowl
MCP server: 447 pay-as-you-go research tools (SEO, ads, SERP) + verified prowl_analyze reports
MCP server for IT hardware parts research: normalize PNs, search listings, get subs/comps.
Read-only MCP server for The Quiet Protocol's engines, benchmarks, proof, and business data.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceBridge the gap between your web crawl and AI language models. With mcp-server-webcrawl, your AI client filters and analyzes web content under your direction or autonomously, extracting insights from your web content. Supports WARC, wget, InterroBot, Katana, and SiteOne crawlers.46Python-
- AlicenseAqualityCmaintenanceMCP server for the Internet Archive's Wayback Machine. Search archived snapshots, extract page text from a specific date, track how a site has changed over time, check if broken links are recoverable, and perform research across Internet Archive collections.63MIT
- AlicenseBqualityNot gradedmaintenanceProvides advanced Google Custom Search functionality, web content extraction, and specialized research tools such as search analytics, multi-site search, and fact checking. Works as an MCP server compatible with any MCP client.10MIT
- AlicenseAqualityAmaintenanceMCP server for searching Swiss court decisions from federal and cantonal courts via entscheidsuche.ch. Enables full-text search, law reference lookup, and filtering by canton, court level, and date without API keys.81MIT