apiwatch

io.github.ReignVentures由 GitHub 帳號 ReignVentures 發布v1.1.0更新於 Oct 28, 2026

Is this model or API still safe to call? Retirement and breaking change dates, with the fix.

已驗證Streamable HTTP可網頁執行開發者工具AI 與機器學習

概覽

AI 產生的概覽

讓助理對照 apiwatch 經審核的廠商變更記錄，查核 AI 模型 ID、API 端點與 MCP 工具的下線、破壞性變更與棄用資訊。

可以做什麼
  • 查詢單一模型 ID，回傳其停用或行為變更日期與因應方式
  • 一次查核多個識別碼：模型 ID、API 端點、MCP 工具名稱、SDK 方法與欄位名稱
  • 依日期列出即將到來的變更，可依天數或廠商篩選
  • 以文字搜尋變更記錄，並讀取單筆記錄的完整內容
  • 列出受監測的公開 MCP 伺服器，查看其工具是否被移除、改名或改動
適用情境
適合在撰寫或審查涉及 AI 模型、API 端點或 MCP 工具的程式碼時使用，也適合依廠商停用日期規劃遷移。在依賴某個公開 MCP 伺服器之前，也可先確認其工具是否已有變動。
執行需求
遠端 Streamable HTTP 端點，免安裝、免帳號、免 API 金鑰；需要能連線至 apiwatch 主機的網路。
安裝前請注意
回答僅來自 apiwatch 經審核的記錄，「無記錄」代表未知而非安全。部分記錄只適用於特定儲存庫或檔案，需留意每筆記錄的適用範圍。查詢內容會傳送給第三方服務，其聲明不保存所收到的內容。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 apiwatch，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.ReignVentures%2Fapiwatch/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": {
    "apiwatch": {
      "type": "http",
      "url": "https://apiwatch.reignventures.co/mcp"
    }
  }
}

README

apiwatch

Know when an API, model, or MCP server you depend on changes, and exactly which lines of your code it breaks.

Vendors retire models, rename MCP tools, and change API behavior on a schedule you don't control. Most of these changes are announced in a changelog nobody on your team reads. apiwatch reads them for you, and its GitHub Action points at the exact lines they land on, before the shutoff date.

  • The feed: reviewed records of breaking, silent, and deprecation changes from OpenAI, Anthropic, Google, Stripe, Twilio, Notion, Atlassian, GitHub, xAI, and more. Every record cites the vendor's own page and is rechecked daily. Browse it at apiwatch.reignventures.co (Atom and JSON Feed included).
  • The scan: finds the call sites each change lands on in your repo. It is token aware (gpt-4 does not match gpt-4o), follows constants, reads notebooks, and separates real uses from entries in model lists.
  • The brief: for each hit, what changed, where, a suggested fix, and acceptance criteria you can hand to a coding agent.
  • The probe: baseline a live JSON endpoint or an MCP server's tools/list and get told when it drifts, no spec required.

What it caught

The feed holds 135 reviewed changes, each checked against the vendor's own page.

  • By vendor: Google 33, Stripe 30, OpenAI 26, Anthropic 17, Notion MCP 8, Atlassian Rovo MCP 4, GitHub MCP Server 4, Notion 4, xAI 4, Twilio 3, Greenhouse 1, MCP specification 1
  • By severity: breaking 100, silent 25, deprecation 10

Next dates (everything in the next 180 days):

GitHub Action

yaml
name: apiwatchon: [push, pull_request]jobs:  scan:    runs-on: ubuntu-latest    permissions: { contents: read, pull-requests: write }    steps:      - uses: actions/checkout@v4      - uses: ReignVentures/apiwatch@v1        with:          fail-on-breaking: "true"   # fail CI when a breaking or silent change touches your code          comment: "true"            # keep one up to date report comment on pull requests          only-changed: "true"       # on PRs, gate only on the files the PR changes

Every run writes a short report to the job summary: which changes touch your code, how severe, whether the date is past or coming, and links to each location. Add SARIF upload to see hits as code scanning annotations:

yaml
      - uses: github/codeql-action/upload-sarif@v3        if: always()        with: { sarif_file: apiwatch.sarif }

Tip: add a schedule: trigger (for example daily) so you hear about new changes even when nobody is pushing.

CLI

bash
pip install git+https://github.com/ReignVentures/apiwatchapiwatch scan . --fail-on-breaking --markdown report.mdapiwatch probe docs "mcp:stdio:npx -y @modelcontextprotocol/server-everything"apiwatch probe docs mcp:https://example.com/mcp --header 'Authorization: Bearer $TOKEN' --fail-on-drift

Scans run locally with no network calls. Your code never leaves your machine or your CI runner.

To leave generated or copied files out of a scan, add a .apiwatchignore file at the repository root: one glob per line (agents/*.json), a trailing / for a directory (generated/), # for comments. Translation folders (locales/, i18n/) are skipped by default.

MCP server

AI agents can ask the feed directly while they work: does this model id or API have a published change, when, and what replaces it? One public URL, no install, no account: https://apiwatch.reignventures.co/mcp (Streamable HTTP, both the 2026-07-28 and the earlier initialize protocol eras).

bash
claude mcp add --transport http apiwatch https://apiwatch.reignventures.co/mcp

Tools: check_model, check_identifiers, list_upcoming, search_changes, get_record. It answers only from the reviewed records the site publishes, stores nothing it is sent, and says "no record" (never "safe") for an id it doesn't know. Setup for other clients: https://apiwatch.reignventures.co/mcp-server

Two more tools, list_watched_mcp_servers and mcp_server_changes, answer from apiwatch's history of observed tool changes on some public MCP servers (https://apiwatch.reignventures.co/mcp-servers/): tools removed or renamed, new required parameters, rewritten descriptions, changed hints. They never return a server's own tool descriptions.

How records are verified

A record is published only after a reviewer that did not write it checks every field against the vendor's primary source, and an automated check confirms the quoted evidence is still on that page. The full rules are in docs/REVIEW.md. If a vendor edits or removes the page, the record comes off the site until it is reviewed again.

Found a change we missed, or a record that's wrong? Open an issue with a link to the vendor's page.

apiwatch Pro

Pro watches every repository in your GitHub organization each night, with nothing to add to your workflows, and opens an issue when a new change lands on your code. $19 per month. See how Pro works.

License

MIT. Built by Reign Ventures LLC.

來源：README.md，提交 30c7bf1

工具

7

check_model

Check an AI model id

Look up one AI model id (for example gpt-4o-2024-05-13 or claude-sonnet-4-5) in apiwatch's reviewed feed of vendor changes before you use it in code. Returns a one line answer (when it stops working, changes behavior, or its deprecation key date) and each matching record with the date, what to do and the replacement, and links to the record and the vendor's source. "No record" means apiwatch has no record for it, not that it is safe. Records that apply only in some repos or files say so in scope.

輸入參數model必填

check_identifiers

Check many identifiers

Look up many identifiers at once before shipping code: model ids, API endpoints (/v1/assistants), MCP tool names, SDK methods or field names. For each one, a one line answer and the ids of matching records, plus the records themselves. A record scoped to some code (for example only repos that depend on a given SDK major version) is returned with that scope spelled out, never as an unconditional hit. Max 100 identifiers.

輸入參數identifiers必填

list_upcoming

List upcoming changes

List the dated changes in apiwatch's feed from today forward, soonest first: model retirements, breaking API changes, silent behavior changes and deprecation key dates. Use it to plan migrations. Optionally limit to the next N days (default 90, max 365) or one vendor. Max 50 records.

輸入參數daysvendor

search_changes

Search changes

Search apiwatch's feed of reviewed vendor changes by text: a case-insensitive match on each record's summary, vendor and signatures (the strings to look for in code). Use it when you know a product or feature name but not the exact id. Newest key date first; default 20 results, max 50.

輸入參數limitquery必填vendor

get_record

Get one record

Get one apiwatch record in full, as published: what changes, what to do, the strings to search for, ids it does not cover, where it applies, the vendor source and when it was reviewed. Use an id from another tool's result.

輸入參數id必填

list_watched_mcp_servers

List watched MCP servers

List the public MCP servers whose tools apiwatch checks every 6 hours (tools/list only, no login, no tool calls), with each one's status (answering, needs a login, not reachable), tool count, last check and last observed change. Use it to see which servers mcp_server_changes can answer about. Returns no tool descriptions.

無需輸入參數。

mcp_server_changes

Recent tool changes on an MCP server

Before you rely on a public MCP server's tools, ask whether apiwatch saw them change: tools removed or renamed, new required parameters, rewritten descriptions, changed safety hints. Give a watched server's slug, URL or registry name. Returns a one line answer and the confirmed changes in the last N days (default 90, max 365), newest first, with flags worth a look. Observations only: no changes is not a statement that a server is safe. Never includes tool description text.

輸入參數daysserver必填

版本歷史

2
  1. v1.1.0最新Oct 26, 2026
  2. v1.0.0Oct 21, 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