qed.bot register

bot.qed由 qed.bot 發布官方發布者v0.3.0更新於 Nov 22, 2026

AI claims in mathematics, graded by evidence. Lean proof checks agents can pay for over x402.

發布方自己的服務:Register.gy

已驗證Streamable HTTP可網頁執行知識與記憶開發者工具AI 與機器學習

概覽

AI 產生的概覽

數學命題與 AI 成果聲明的唯讀登錄庫,讓助理檢索結果、形式化、目標與證明檢查結論。

可以做什麼
  • 檢索並讀取命題,含形式化狀態、獎金與已記錄的 AI 工作
  • 檢視形式化宣告、其公理宣告以及由誰核驗
  • 比較各登錄庫對同一結果的立場與分歧
  • 瀏覽目標、紀錄歷史、合集、未解問題與基準集
  • 檢視揭露評級、物件檢查器與各 AI 系統的戰績
  • 提交 Lean 檔案付費重建並讀取檢查結論
適用情境
適合在需要確認某個數學結果已被形式化到什麼程度、記錄了多少 AI 工作、不同登錄庫如何評判同一結論時使用,也適合需要對證明檔案做獨立 Lean 重建時使用。
執行需求
遠端 streamable HTTP 端點 qed.bot;讀取類工具無需帳號。付費檢查需在網站購買額度,並以 Authorization Bearer 標頭送出 API 金鑰;私有檢查需同一金鑰。
安裝前請注意
check_proof 會消耗額度(每次一額度,私有檢查兩額度),並需要妥善保管的 API 金鑰。此端點只讀取登錄庫;驗證或提交證明與構造會執行作者程式碼,僅由本機伺服器提供,此遠端端點不提供。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 qed.bot register,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/bot.qed%2Fregister/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": {
    "qed-bot-register": {
      "type": "http",
      "url": "https://qed.bot/mcp"
    }
  }
}

工具

20

search_statements

Search the register of mathematical statements. Returns summaries with formal status, prizes, and how much AI work is recorded against each.

輸入參數limitqueryhas_prizecollectionhas_ai_claimformal_statusverified_only

get_statement

Everything the register holds about one statement: formalisations, cited proofs, recompilation results, graded claims, relations and aliases.

輸入參數id必填

compare_registries

Results recorded by more than one registry, and how the positions differ. A disagreement is a difference of judgement about who did the mathematics; a gap is one registry holding stronger evidence than another; a lead is a proof somebody else records that this register has not rebuilt.

輸入參數limitfindingregistry

registry_positions

Every registry's position on one result, this register included, with each one's own label preserved beside the bucket it was placed in.

輸入參數id必填

search_formalisations

Every formalization.yaml the register has read, with what each declares about its own axioms, unproved goals, assumed literature, automation and review, and who has checked it: this register, Palomar, or nobody independent of its authors.

輸入參數flaglimitquerymethodreviewheld_onlychecked_by

get_formalisation

One declaration in full: its sources, related formalisations, main results, the results held here it joins and on what key, Palomar's registrations of it, and any contradiction between what it declares and what a rebuild found.

輸入參數id必填

disclosure

Announcements of results, graded for how much of what they claim can be inspected: whether each result is named, written up, and published with a checkable artifact.

輸入參數org

list_evaluators

The object checkers available, what each establishes, what an artifact must look like, and which targets have been checked with it. Read this before calling submit_construction, which only the local server offers.

無需輸入參數。

search_targets

Search the targets: quantities with a direction and a record history, where AI work moved a bound or produced a better object rather than settling a statement.

輸入參數limitquerystandingcollectionchecked_onlyheld_by_machine

get_target

Everything the register holds about one target: the record history with who held each value and when, the graded claims, and any check run here.

輸入參數id必填

list_collections

The collections held, with counts of statements, formalisations and AI activity.

無需輸入參數。

open_problems

Statements not recorded as solved, most valuable prize first. Use formalised_only to restrict to those a prover could attempt directly.

輸入參數limitmin_prizecollectionformalised_only

list_systems

Track record per AI system, computed from recorded claims. Not a benchmark: which statements attract attention has little to do with how hard they are.

輸入參數namelimit

benchmarks

Benchmarks as sets of statements, with how many members already carry public AI work: a contamination figure rather than a study.

輸入參數include_members

check_proof

Have qed.bot rebuild a Lean file at its commit and read what every theorem in it depends on, under the admission rules every stranger's file is held to. Costs one credit, or two for a private check; buy credits and make an API key at https://qed.bot/check, and send it as Authorization: Bearer qed_.... Where agent payments are open, an agent with no key may instead pay for the one check in USDC over x402: called without a key, it answers with the payment asked for, and a retry carrying the payment in params._meta["x402/payment"] queues the check. Returns at once with the check queued; call get_check until it is done, which takes minutes. A file already checked in public is returned free, and a check that stops on qed.bot's side is refunded.

輸入參數url必填privatetheorem

get_check

A paid check's status and verdict, by the id check_proof returned. Needs the same API key for a private check, or, for one paid over x402, the token check_proof returned.

輸入參數id必填token

submit_lead

Send qed.bot a link it should look at: a paper, a repository, a post or an announcement of AI work in mathematics, with an optional note, and no record to choose. A link already waiting as a lead is returned rather than sent twice. Needs an API key, free from https://qed.bot/check (no credits are needed to submit), sent as Authorization: Bearer qed_.... It goes to review: nothing in the register changes until an administrator accepts it in public, and grades are never bought. Returns the submission's id; poll get_submission with it.

輸入參數url必填notetitle

submit_claim

Add a claim of AI work to a record the register holds, with the systems and people behind it, when it was done, what it achieved and its sources. A Lean proof linked at its commit is queued for qed.bot's checker. Autonomy is recorded as declared until an administrator grades it from the sources. Needs an API key, free from https://qed.bot/check (no credits are needed to submit), sent as Authorization: Bearer qed_.... It goes to review: nothing in the register changes until an administrator accepts it in public, and grades are never bought. Returns the submission's id; poll get_submission with it.

輸入參數datenode必填titlehumansdetailsoutcome必填sources必填systems必填theoremproof_url

flag_record

Ask for a correction to a record: say what is wrong, with sources. Sent as a correction. Needs an API key, free from https://qed.bot/check (no credits are needed to submit), sent as Authorization: Bearer qed_.... It goes to review: nothing in the register changes until an administrator accepts it in public, and grades are never bought. Returns the submission's id; poll get_submission with it.

輸入參數node必填what必填titlesources必填claim_id

get_submission

Where a submission stands, by its id: stage received, in_review, accepted or declined, the review's note (the reason for a decline), whether the register now carries it, and its public thread. Needs no key.

輸入參數id必填

版本歷史

2
  1. v0.3.0最新Nov 22, 2026
  2. v0.2.0Nov 21, 2026

評分

已滿足適用檢查項的 72%

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

    請在註冊中心條目中發布圖示。

  • 文件15 分

    請宣告包含 README 的 GitHub 儲存庫,或在 OAuth 受保護資源後設資料中提供 resource_documentation。

  • 授權條款不適用

    託管服務按其提供方的條款使用,因此這裡不要求授權條款。此檢查項不計入百分比。

其他 Register 整合