Intelligence Aeternum

io.github.codex-curator由 GitHub 帳號 codex-curator 發布v1.0.0更新於 Oct 24, 2026

AI training dataset marketplace: 2M+ museum artworks with Golden Codex enrichment

已驗證Streamable HTTP可網頁執行資料與分析商業與電商媒體與設計

概覽

AI 產生的概覽

遠端 MCP 市集,可搜尋 200 萬件以上博物館藝術品並購買 AI 訓練中繼資料,提供免費搜尋與付費增強、交付工具。

可以做什麼
搜尋與瀏覽藝術品及資料集目錄,包含資料集預覽與價格查詢,並提供代理工作流程指南、增強層級清單與已提交影像的增強狀態查詢。取得單件藝術品的中繼資料與影像,可選免費的人工來源中繼資料或付費的 VLM 深度分析中繼資料。為資料集產生法規合規清單。提交自己的影像進行增強,並建立最多 1 萬件藝術品的隨選交付訂單。透過 x402 或 Stripe 發起資料集購買。
適用情境
當助理需要公有領域博物館藝術品資料或已增強的 AI 訓練中繼資料,或需要為訓練資料集取得來源與合規文件時適用。若只需查詢來源博物館 API 已免費提供的影像,則沒有必要使用。
執行需求
遠端 streamable HTTP 端點;未宣告本機套件、API 金鑰或環境變數。付費工具需要付款:代理使用 Base L2 上的 x402 USDC,人工買家使用 Stripe 結帳。需要能存取該託管服務的網路。
安裝前請注意
多個工具需要付費:付費中繼資料、影像增強、藝術品交付與資料集購買,並有大量折扣與上線優惠價。代理付款使用 Base L2 上的 USDC,需要錢包以及攜帶交易證明的 X-PAYMENT 標頭。購買與交付訂單會建立訂單並花費資金。提交的影像會傳送給服務方進行分析,增強中繼資料由營運方商業授權。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Intelligence Aeternum,將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable,透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式,或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南,先向你展示發布者、端點和登入方式,經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.codex-curator%2Fintelligence-aeternum-mcp/install.md,並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前,先告訴我它的發布者、端點和登入方式,等我確認。不要把 API 金鑰寫進命令或我會提交的檔案;問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "intelligence-aeternum": {
      "type": "http",
      "url": "https://data-portal-172867820131.us-west1.run.app/mcp"
    }
  }
}

README

Intelligence Aeternum MCP

AI training dataset marketplace exposed as MCP servers. 2M+ museum artworks across 7 world-class institutions with on-demand 111-field Golden Codex AI enrichment. AI agents pay with x402 USDC on Base L2. Human buyers pay with Stripe.

[Research] [Dataset] [License]


Quick Start

Connect to the MCP endpoint with any MCP-compatible client:

json
{
  "mcpServers": {
    "intelligence-aeternum": {
      "url": "https://data-portal-172867820131.us-west1.run.app/mcp",
      "transport": "streamable-http"
    }
  }
}

Or use the MonetizedMCP broker (x402 payment orchestration):

json
{
  "mcpServers": {
    "fluora-mcp": {
      "url": "https://fluora-mcp-172867820131.us-west1.run.app/mcp",
      "transport": "streamable-http"
    }
  }
}

Architecture

AI Agent (Claude, GPT, etc.)
    |
    +---> Fluora MCP (MonetizedMCP broker)         [TypeScript/Express]
    |     3 tools: price-listing, payment-methods, make-purchase
    |     Orchestrates x402 USDC payments via MonetizedMCP SDK
    |     Proxies to Data Portal with payment proof
    |
    +---> Data Portal MCP (FastMCP content server)  [Python/FastAPI]
          13 tools: search, metadata, oracle, enrichment, compliance...
          Serves 2M+ artworks from Firestore manifest
          On-demand VLM enrichment via Gemini

Human Buyer (Browser)
    |
    +---> Data Portal REST API / golden-codex.com
          Stripe checkout for dataset purchases and enrichment
          Same artifact delivery pipeline as x402

Services

Data Portal (13 MCP Tools)

ToolPaid?Description
search_alexandriaFreeSearch 2M+ museum artworks across 7 institutions
get_curated_metadataFree (5/day)Human_Standard metadata + image (500-1,200 tokens)
get_oracle_metadata$0.16 USDCHybrid_Premium 111-field VLM deep analysis + image
get_compliance_manifestFreeAB 2013 + EU AI Act Article 53 compliance manifests
search_datasetsFreeBrowse 7 museum dataset catalogs
preview_datasetFreeSample images from a dataset
get_pricingFreeCalculate pricing with volume discounts
get_agent_guideFreeComplete API workflow documentation
list_enrichment_tiersFreeAvailable enrichment tiers and pricing
get_enrichment_statusFreePoll enrichment job status
purchase_datasetPaidInitiate batch dataset purchase
enrich_agent_imageFrom $0.16Submit YOUR image for Golden Codex enrichment
deliver_artifactsFrom $0.04On-demand artifact delivery

Fluora MCP (MonetizedMCP Broker)

ToolDescription
price-listingBrowse 7 purchasable data products with pricing
payment-methodsGet accepted payment methods (x402 USDC for agents, Stripe for humans)
make-purchasePurchase data with x402 USDC micropayment

Pipeline Tools (Coming to MCP)

ToolDescriptionPrice
SD 3.5 Large + T5-XXLImage generation on NVIDIA L4 GPU with LoRA supportTBD
ESRGAN x4 UpscalerReal-ESRGAN super-resolution (1024px to 4096px in ~1.15s)$0.10
Nova (Metadata Creation)111-field Golden Codex VLM analysis via Gemini$0.20
Atlas (Metadata Infusion)XMP/IPTC/C2PA embedding + SHA-256 Soulmark + hash registry$0.10
Aegis (Verification)"Shazam for Art" — perceptual hash provenance lookupFree
Archivus (Arweave Storage)Permanent 200+ year storage — pay USDC, no AR neededTBD
Mintra (NFT Minting)Polygon NFT with full Golden Codex metadata on-chainTBD

Museums

InstitutionArtworksLicense
Metropolitan Museum of Art375,000CC0
Rijksmuseum709,000CC0
Smithsonian Institution185,000CC0
National Gallery of Art130,000CC0
Art Institute of Chicago120,000CC0
Cleveland Museum of Art61,000CC0
Paris Collections (Louvre, Orsay, Rodin)45,000CC0

All source images are CC0/public domain. Enrichment metadata is commercially licensed by Metavolve Labs.


Payment

Two payment paths -- one marketplace, two buyer types:

  • AI agents pay with x402 USDC micropayments on Base L2. No API keys, no subscriptions -- agents pay autonomously per request.
  • Human buyers pay with Stripe checkout. Credit/debit cards, Apple Pay, Google Pay -- standard web checkout for dataset purchases and enrichment orders.

x402 Flow (AI Agents)

  1. Call a paid tool without payment
  2. Receive HTTP 402 with x402 payment envelope
  3. Execute USDC transfer on Base L2
  4. Re-call with X-PAYMENT header containing transaction proof
  5. Receive data

Wallet: 0xFE141943a93c184606F3060103D975662327063B

Stripe Flow (Human Buyers)

  1. Select data products or enrichment services
  2. Complete Stripe checkout (cards, Apple Pay, Google Pay)
  3. Receive artifacts via the same delivery pipeline

Genesis Epoch (Launch Pricing)

20% off all prices for 90 days (started Feb 23, 2026).

Volume discounts auto-apply per wallet:

  • 100+ records: 25% off
  • 500+ records: 37% off
  • 2000+ records: 50% off

Discovery

EndpointURL
Data Portal MCPhttps://data-portal-172867820131.us-west1.run.app/mcp
Fluora MCPhttps://fluora-mcp-172867820131.us-west1.run.app/mcp
OpenAPI Docshttps://data-portal-172867820131.us-west1.run.app/docs
llms.txt (Data Portal)https://data-portal-172867820131.us-west1.run.app/llms.txt
llms.txt (Fluora)https://fluora-mcp-172867820131.us-west1.run.app/llms.txt
.well-known/mcp.jsonhttps://data-portal-172867820131.us-west1.run.app/.well-known/mcp.json
Schema (Fluora)https://fluora-mcp-172867820131.us-west1.run.app/mcp/schema
Health (Data Portal)https://data-portal-172867820131.us-west1.run.app/health
Health (Fluora)https://fluora-mcp-172867820131.us-west1.run.app/health

Research

Our preprint The Density Imperative reports that dense metadata measurably affects VLM capability:

  • +25.5% improvement with 111-field Golden Codex enrichment
  • -54.4% degradation with sparse 10-20 word captions
  • +160% semantic coverage vs raw captions

Paper: DOI: 10.5281/zenodo.18667735


Directory Structure

intelligence-aeternum-mcp/
+-- fluora-mcp/              # MonetizedMCP broker (TypeScript/Express)
|   +-- src/
|   |   +-- main.ts          # Entrypoint
|   |   +-- server/
|   |   |   +-- server.ts    # MCP server + HTTP endpoints
|   |   |   +-- purchasableItems.ts  # 7 product catalog
|   |   |   +-- paymentMethods.ts    # USDC on Base L2
|   |   +-- config/
|   |       +-- config.ts    # Environment config
|   +-- Dockerfile
|   +-- package.json
|   +-- deploy.sh
|
+-- data-portal/             # FastMCP content server (Python/FastAPI)
|   +-- main.py              # FastAPI app + MCP mount
|   +-- mcp_server.py        # 13 MCP tool definitions
|   +-- auth.py              # x402 payment verification + rate limiting
|   +-- pricing.py           # Pricing tiers + Genesis Epoch
|   +-- compliance.py        # AB 2013 + EU AI Act manifests
|   +-- volume_tracker.py    # Per-wallet volume discounts
|   +-- image_fetcher.py     # Museum API integration
|   +-- routes/
|   |   +-- agent.py         # Agent API endpoints
|   |   +-- reader.py        # Verilian Reader (decode XMP metadata)
|   |   +-- catalog.py       # Dataset catalog
|   |   +-- orders.py        # Stripe + x402 orders
|   |   +-- enrich.py        # Enrichment-as-a-Service
|   |   +-- deliver.py       # On-demand delivery
|   |   +-- admin.py         # Admin utilities
|   +-- Dockerfile
|   +-- requirements.txt
|   +-- deploy.sh
|
+-- glama.json               # Glama.ai MCP directory manifest
+-- README.md                # This file

Compliance

Every purchase auto-generates provenance manifests at no additional cost:

  • AB 2013 (California AI Training Data Transparency Act)
  • EU AI Act Article 53 (Training data documentation requirements)

Enterprise

Full dataset access with compliance manifests starting at $8,000.

Contact: [email protected]


Important: What You're Paying For

Pricing applies to structured AI training metadata only -- not the source images. All 2M+ source artworks are CC0/public domain from partner museums and freely available from their original APIs. What Intelligence Aeternum provides is:

  • Golden Codex enrichment: 111-field NEST metadata generated by multi-model VLM analysis (Gemini + Claude + GPT-4o)
  • Compliance manifests: Auto-generated AB 2013 + EU AI Act Article 53 documentation
  • Pipeline services: Upscaling, metadata infusion, C2PA signing, hash registration, permanent storage
  • Structured datasets: Pre-enriched collections optimized for AI training workflows

License

Source code in this repository is provided for reference and transparency. The enrichment metadata, Golden Codex schema, and pipeline services are proprietary to Metavolve Labs, Inc. Source images from museums are CC0/public domain.


Links


Metavolve Labs, Inc. | San Francisco, California "Synthetic Data is not the problem. Synthetic Garbage is."

來源:README.md,提交 6ac0b77

工具

13

search_alexandria

Search Alexandria

FREE — Search 2M+ museum artworks in the Alexandria Aeternum catalog. Searches the Firestore manifest (primary) with Vertex AI fallback. Returns artifact IDs, titles, artists, dates, and classification. Use get_curated_metadata() for Human_Standard metadata + image (5 free/day). Use get_oracle_metadata() for Hybrid_Premium VLM deep analysis + image.

輸入參數limitquery必填museum

get_curated_metadata

Get Curated Metadata

FREE (5/day) — Get Human_Standard metadata + image for an artifact. Returns 500-1,200 tokens of 100% human-sourced metadata PLUS a signed image download URL. Every response is a package: metadata + image together. Zero synthetic content. Sources: Museum API + Wikipedia + Wikidata + Getty ULAN. After 5 free requests per day, requires $0.05 USDC via x402.

輸入參數artifact_id必填

get_oracle_metadata

Get Oracle Metadata

PAID — Get Hybrid_Premium metadata + image with VLM deep visual analysis. Returns 2,000-6,000 tokens including everything in Human_Standard PLUS: visual_analysis, emotional_and_thematic_journey, deep symbolism, and archetypal analysis. Image download URL included in response.

輸入參數artifact_id必填

get_compliance_manifest

Get Compliance Manifest

FREE — Get auto-generated regulatory compliance manifests for a dataset. Returns AB 2013 (California) and/or EU AI Act Article 53 provenance manifests. "Buy from us, get instant California + EU AI compliance."

輸入參數dataset_id必填regulation

get_agent_guide

Get Agent Guide

FREE — Get complete API workflow documentation for agents. Returns the full agent guide with endpoints, pricing, custom fields schema, volume discounts, and recommended workflow.

無需輸入參數。

search_datasets

Search Datasets

FREE — Search the Alexandria Aeternum dataset catalog (7 museums, 1.6M+ images).

輸入參數querydomainmin_images

preview_dataset

Preview Dataset

FREE (10/day) — Preview sample images from a dataset.

輸入參數limitdataset_id必填

get_pricing

Get Pricing

FREE — Get pricing information for a dataset purchase.

輸入參數quantitydataset_id必填

purchase_dataset

Purchase Dataset

Initiate a dataset purchase (x402 or Stripe).

輸入參數quantity必填dataset_id必填payment_method

list_enrichment_tiers

List Enrichment Tiers

FREE — List available enrichment tiers for agent-submitted images. NEW: Submit YOUR images via POST /agent/enrich with custom fields. Your metadata is merged with Oracle analysis. Submitter values take priority.

無需輸入參數。

enrich_agent_image

Enrich Agent Image

Submit YOUR image for Golden Codex enrichment (Enrichment-as-a-Service).

輸入參數tierimage_url必填callback_urlcustom_fields

deliver_artifacts

Deliver Artifacts

Create an on-demand delivery order for specific manifest artifacts. Artifacts are fetched from museum APIs, optimized, enriched (Nova + Atlas), and bundled into a single downloadable archive. Supports up to 10,000 artifacts per order with automatic volume discounts.

輸入參數tierformatartifact_ids必填

get_enrichment_status

Get Enrichment Status

Poll the status of a Golden Codex enrichment job.

輸入參數job_id必填

版本歷史

1
  1. v1.0.0最新Sep 16, 2026

評分

已滿足適用檢查項的 90%

該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分

    請在儲存庫或註冊中心條目中宣告授權條款。