Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
作者提供的说明(英文原文)
EU cookie-consent audit: trackers before consent, Reject all, Consent Mode v2, TCF; evidence + fixes
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
适用于网页版、桌面版和手机版 Claude:
https://api.bannerproof.app/mcp在终端中运行这条命令
claude mcp add --transport http bannerproof https://api.bannerproof.app/mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"bannerproof": {
"url": "https://api.bannerproof.app/mcp"
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"bannerproof": {
"type": "http",
"url": "https://api.bannerproof.app/mcp"
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
ChatGPT 只能连接互联网上的服务器:
https://api.bannerproof.app/mcp菜单名称可能因套餐和版本不同而略有差异。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
服务器要求先登录,才会列出它的工具。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月8日
注册名称已通过域名 bannerproof.app 验证(注册中心会核实域名所有权)
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
在 517 毫秒内响应(HTTP 401)
声明支持 OAuth 登录,你的 AI 应用可以带你去登录
尚未检查;我们会在检查端点时读取工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 3 项适用的基础检查中通过了 3 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在作者的云端运行。通过互联网连接,无需安装,但你发送的内容会经过作者的服务器。
GitHub 仓库最后更新于–,最新版本(1.1.1)于8天前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。它通过 URL 连接,所以接受远程 MCP 服务器的应用也可以使用,例如开启开发者模式的 ChatGPT。
同一分类中有:Spec Driven Development, Remote Desktop Commander, alalanet — check before you pay
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Check packages, MCP servers and x402 endpoints before agents install or pay. Signed.
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Bannerproof checks a website the way a real first-time visitor from the EU sees it, and reports:
ad_storage, analytics_storage, ad_user_data, ad_personalization);Every finding comes with the exact request as evidence, a concrete fix (13 languages) and the legal reference (ePrivacy Art. 5(3), GDPR, national law). Works with any consent platform — Cookiebot, Usercentrics, OneTrust, Complianz, CookieYes, Borlabs, iubenda or a custom banner.
This repository holds the public pieces: the MCP server listing, connection examples and a GitHub Action. The service itself runs at bannerproof.app.
Remote MCP server, Streamable HTTP, OAuth 2.0:
https://api.bannerproof.app/mcp
A free quick check works without an account (https://api.bannerproof.app/mcp/free). Full findings, monitoring, audits and evidence exports use a Bannerproof account — sign in with OAuth when your client asks, or use an API token from Settings.
| Client | How to connect |
|---|---|
| Claude (claude.ai / Desktop) | Settings → Connectors → find Bannerproof in the directory, or add a custom connector with the URL above |
| Claude Code | claude mcp add --transport http bannerproof https://api.bannerproof.app/mcp |
| Cursor | see examples/cursor-mcp.json |
| Cline | see llms-install.md |
| Other clients | add an MCP server with the URL above (OAuth), or send Authorization: Bearer |
Step-by-step guide: bannerproof.app/help/connect-claude-chatgpt
Tools
| Tool | What it does | Account |
|---|---|---|
quick_check | yes/no + score for any public site | free |
explain_finding | what a finding means, why it matters, how to fix it, legal basis | free |
get_capabilities | what the server checks and what each tool does | free |
get_findings | full findings with evidence, fixes and legal references | yes |
scan_domain | fresh scan of a monitored domain | yes |
compare_scans | what changed since the previous scan | yes |
get_evidence | evidence pack (every request with phase and time, cookies, consent signals) as CSV/JSON | yes |
get_report_link | private report link to send to a client or developer | yes |
audit_domain / audit_domains | audit any site, or up to 10 at once (1 audit credit each; failed scans are refunded) | yes |
get_audits_status | status of several audits at once | yes |
list_domains, add_domain, scan_history, account_info, get_upgrade_link | account helpers | yes |
The server is listed in the official MCP Registry as app.bannerproof/bannerproof (server.json).
Fail a pipeline when a release breaks cookie consent. The domain must be monitored in your Bannerproof account; create an API token in Settings and store it as a secret.
- uses: bannerproof/bannerproof@v1
with: domain: example.com api-token: ${{ secrets.BANNERPROOF_TOKEN }} fail-on: new # "new" = only issues this release added (default), "any" = any issue
Full workflow: examples/github-workflow.yml.
curl -fsS -X POST https://api.bannerproof.app/api/ci/check \
-H "X-Api-Token: $BANNERPROOF_TOKEN" -H "Content-Type: application/json" \ -d '{"domain":"example.com","failOn":"new"}'
| HTTP status | Meaning |
|---|---|
| 200 | pass |
| 422 | fail — the JSON body lists new_issues |
| 202 | still scanning — poll the poll URL from the body |
| 429 | the domain was scanned less than 5 minutes ago (Retry-After header) |
| 502 | the site couldn't be scanned |
In Settings you can also add a webhook: after every scan of a monitored domain Bannerproof POSTs a scan.completed event (score, new and resolved issues) signed with HMAC-SHA256 — X-Bannerproof-Signature: sha256= . ")>.
Bannerproof is an automated technical check of what a site sends at scan time — not legal advice.
Support: support@bannerproof.app