AnchorRegistry

io.github.AnchorRegistry由 GitHub 帳號 AnchorRegistry 發布v0.1.0更新於 Oct 24, 2026

On-chain provenance lookup for AnchorRegistry. Resolve AR-IDs, hashes, and full trees. Authless.

已驗證Streamable HTTP可網頁執行開發者工具安全與監控

概覽

AI 產生的概覽

AnchorRegistry 是遠端、免驗證的 MCP 伺服器,讓助理依 AR-ID、SHA-256 雜湊或完整樹查詢成品的鏈上來源紀錄。

可以做什麼
將 AR-ID 解析為完整來源紀錄,包括誰錨定了該成品、何時錨定、成品類型,以及樹是否已封存或撤銷,適用於在 README、論文、模型卡或網站中遇到 SPDX-Anchor 或 DAPX-Anchor 標籤時。依 SHA-256 清單雜湊解析成品,檢查是否已被錨定,回傳 AR-ID 與紀錄,或明確回傳未錨定。解析 AR-ID 的完整來源樹,回傳所有錨點及其關係、類型、清單雜湊與時間戳,用於追蹤多成品來源鏈。
適用情境
當你需要確認軟體成品、模型、資料集或論文的來源與出處時使用,尤其是在遇到 AR-ID 或錨定標籤、想了解它是什麼以及由誰錨定時。也適合核驗多成品來源鏈,例如以論文為根節點並關聯其他成品的情況。
執行需求
遠端 MCP 端點;未宣告驗證、API 金鑰或環境變數。用戶端需支援 streamable HTTP 傳輸。工具呼叫會代理到 AnchorRegistry API,因此需要能存取該服務的網路。
安裝前請注意
此伺服器被描述為輕量代理,每次工具呼叫都會轉送到 AnchorRegistry API 並原樣回傳 JSON,因此你的查詢會傳送給該第三方。依 README 說明,它只做唯讀查詢,不儲存憑證,請求之間不保留狀態。結果反映 AnchorRegistry 自身的紀錄,應視為該服務的聲明。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 AnchorRegistry,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.AnchorRegistry%2Far-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": {
    "anchorregistry": {
      "type": "http",
      "url": "https://mcp.anchorregistry.ai/mcp"
    }
  }
}

README

ar-mcp

The AnchorRegistry MCP server. Resolve any AR-ID, manifest hash, or provenance tree from any MCP-compatible AI client.

[License: MIT] [MCP] [Cloudflare Workers] [smithery badge] [arXiv]

Endpoint: https://mcp.anchorregistry.ai/mcp

Install

Claude Desktop

Settings → Integrations → Add server. URL: https://mcp.anchorregistry.ai/mcp

Claude Code

bash
claude mcp add --transport http anchorregistry https://mcp.anchorregistry.ai/mcp

Cursor

Settings → MCP → Add new MCP server. Name: anchorregistry · Type: http · URL: https://mcp.anchorregistry.ai/mcp

Tools

ar_verify_arid

Resolves an AR-ID to its full provenance record.

"What's at AR-2026-qnPOJ1z?"

ar_verify_by_hash

Resolves an artifact by SHA-256 manifest hash.

ar_resolve_tree

Returns the full provenance tree for any AR-ID.

How it works

ar-mcp is a thin proxy. Every tool call hits api.anchorregistry.ai and returns the JSON unchanged. No data is logged, no credentials are stored, no state is held between requests.

The full surface is documented at:

Roadmap

v0.2 will add ar_check_balance and ar_register_artifact once the MCP credential-handling pattern stabilizes across clients. ar_seal_tree is deferred indefinitely (irreversible action; needs robust user-confirmation primitives the protocol does not yet provide). The deferred tools are fully implemented and live in src/future-tools.ts — enabling them is a one-line flip per tool.

Develop

bash
npm installnpm test           # vitest, hits live api.anchorregistry.ainpm run type-checknpm run dev        # wrangler dev — http://localhost:8787npm run deploy     # wrangler deploy

Verify locally with the MCP inspector:

bash
npx @modelcontextprotocol/inspector http://localhost:8787/mcp

License

MIT.

See also

來源:README.md,提交 1762fbf

工具

3

ar_verify_arid

Resolve an AnchorRegistry AR-ID to its full provenance record. Use this when you encounter an SPDX-Anchor or DAPX-Anchor tag (e.g. in a README, paper, model card, or website) and need to confirm what the artifact is, who anchored it, when, what type it is, and whether the tree has been sealed, retracted, voided, or affirmed. Free, public, no auth required.

輸入參數ar_id必填

ar_verify_by_hash

Resolve an artifact by its SHA-256 manifest hash. Use this when you have the artifact itself but no AR-ID, and want to check whether it has been anchored. Returns the AR-ID and full provenance record if found, or a clear "not anchored" response otherwise.

輸入參數manifest_hash必填

ar_resolve_tree

Resolve the full provenance tree for an AR-ID. Returns every anchor in the tree (root + all descendants) with their relationships, types, manifest hashes, and timestamps. Use this when you need to understand a multi-artifact provenance chain — for example, a research paper anchored as the root with its training dataset and model weights as children.

輸入參數ar_id必填

版本歷史

1
  1. v0.1.0最新Sep 16, 2026

評分

已滿足適用檢查項的 100%

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