mcpField
A
agent-ready-cloudflare

agent-ready-cloudflare

github.com/fabricioctelles开发12387 次安装2 天前更新
代码仓库

关于 agent-ready-cloudflare

agent-ready-cloudflare 是用于扫描网站 AI Agent 就绪度并生成检查报告和改进提示词的技能。它检查发现机制、Agent 可读内容、机器人访问控制、API 与 MCP 发现、身份验证及可选的商业协议。

主要功能

  1. 01通过 isitagentready.com API 扫描网站
  2. 02评估 Agent 发现机制和可读内容
  3. 03检查 API、身份验证及 MCP 元数据
  4. 04为未通过或中性检查生成提示词
  5. 05支持多域名扫描并逐步保存结果

适用场景

  1. 01启用 Agent 访问前审计网站
  2. 02找出缺失的发现机制或 MCP 元数据
  3. 03为未通过的检查生成实施提示词

MCP 相关

扫描器也以 MCP 服务形式提供，地址为 https://isitagentready.com/mcp。MCP 客户端可调用 scan_site 工具并传入 URL。

安装

复制下面的命令，在终端中运行。

$npx skills add https://github.com/fabricioctelles/skills --skill agent-ready-cloudflare

安装后在 Claude Code 中输入/agent-ready-cloudflare确认已加载。

附带文件README.md

详细介绍

Agent Ready — Cloudflare Scanner

审核任意网站是否适用于 AI Agent，生成可执行的修复提示词，并实施改进以提高 Agent 就绪度评分。

0. 操作流程

每次激活此技能时，都请遵循以下流程：

Step 1 — 获取域名

如果用户未提供域名，请询问：

你想扫描 isitagentready.com 上的哪个域名？

Step 2 — 通过 API 扫描
curl -s -X POST 'https://isitagentready.com/api/scan' \
  -H 'Content-Type: application/json' \
  -H 'User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/145.0.0.0 Safari/537.36' \
  -H 'Referer: https://isitagentready.com/' \
  -H 'Origin: https://isitagentready.com' \
  -d '{"url":"https://DOMAIN/","enabledChecks":["robotsTxt","sitemap","linkHeaders","dnsAid","markdownNegotiation","robotsTxtAiRules","contentSignals","webBotAuth","apiCatalog","oauthDiscovery","oauthProtectedResource","authMd","mcpServerCard","a2aAgentCard","agentSkills","webMcp","ard","personalAgentProtocol","x402","mpp","ucp","acp","ap2"]}'

将 DOMAIN 替换为目标域名。

Step 3 — 生成 Markdown 报告

使用 API 响应，按以下结构生成报告：

# Agent Ready Scan — {domain}

> **Score:** Level {level} — {levelName}
> **Scanned:** {scannedAt}
> **Link:** [View online](https://isitagentready.com/{domain})

## Summary

| Category | Score |
|----------|-------|
| Discoverability | {passed}/{total} |
| Content | {passed}/{total} |
| Bot Access Control | {passed}/{total} |
| API, Auth, MCP & Skill Discovery | {passed}/{total} |
| Commerce (Optional) | {passed}/{total} |

## Details

### Discoverability ({passed}/{total})

- ✅ **robots.txt** — {message}
- ❌ **sitemap.xml** — {message}
  (... for each check ...)

### (... repeat for each category ...)

## 🔧 How to Implement — Agent Prompts

(... one block per failing check, see Step 4 ...)

## Next Level

**Level {nextLevel.target} — {nextLevel.name}**

To reach the next level, implement:
- {nextLevel.requirements[].description}
Step 4 — 生成“如何实施”提示词

对于每项 status: "fail" 或 status: "neutral" 的检查，使用下方第 8 节中的提示词模板生成提示词块。提示词包含：

  1. 模板中的目标和修复方法（针对每项检查固定）
  2. API 响应中的问题（checks.{category}.{check}.message）
  3. 指向子技能的技能 URL
  4. 相关 RFC/规范的文档链接

将每条提示词格式化为围栏代码块，供用户复制粘贴到编码 Agent 中：

#### ❌ {check name}

**Issue:** {message from API}

```
Goal: {goal from template}

Issue: {message from API}

Fix: {fix from template}

Skill: https://isitagentready.com/.well-known/agent-skills/{skill-folder}/SKILL.md

Docs: {docs URLs from template}
```

> 📖 Reference: [{skill-folder}/SKILL.md]({skill-folder}/SKILL.md)
Step 5 — 交付

向用户提供完整的 Markdown 报告。如果用户想将其保存为文件，请写入指定路径。

1. 检查内容

可发现性（4 项检查）
检查项API 键通过标准
robots.txtrobotsTxt返回 200，且 text/plain 内容至少包含一条 User-agent 指令
sitemap.xmlsitemap/sitemap.xml 返回有效 XML，或 robots.txt 中存在 Sitemap 指令
Link 标头linkHeaders主页包含带有对 Agent 有用的关系（service-desc、api-catalog 等）的 Link 标头
DNS-AIDdnsAid通过 DNS-over-HTTPS 在 _agents 命名空间下找到 SVCB/HTTPS 记录（优先使用 Cloudflare，失败时回退到 Google）
内容（1 项检查）
检查项API 键通过标准
面向 Agent 的 MarkdownmarkdownNegotiationAccept: text/markdown 返回 Content-Type: text/markdown
机器人访问控制（3 项检查）
检查项API 键通过标准
AI 机器人规则robotsTxtAiRulesrobots.txt 包含已知 AI 机器人的 User-agent 条目（GPTBot、Claude-Web、Google-Extended 等）
内容信号contentSignalsrobots.txt 包含 Content-Signal 指令，涵盖 ai-train/search/ai-input
Web Bot AuthwebBotAuth/.well-known/http-message-signatures-directory 存在且包含有效的 JWKS（仅供参考——neutral 不影响评分）
API、身份验证、MCP 和技能发现（10 项检查）
检查项API 密钥通过标准
API 目录apiCatalog/.well-known/api-catalog 返回有效的 linkset+json，其中包含 API 条目
OAuth/OIDCoauthDiscovery/.well-known/openid-configuration 或 oauth-authorization-server 包含有效的 OAuth 元数据
OAuth 受保护资源oauthProtectedResource/.well-known/oauth-protected-resource 包含 resource 和 authorization_servers
Auth.mdauthMd/auth.md 存在，且包含有效的 H1 标题，其中含有 "auth.md"；可选提供 PRM + AS 元数据
MCP 服务卡片mcpServerCard/.well-known/mcp/server-card.json、server-cards.json 或 mcp.json 中存在有效卡片，且包含 serverInfo.name
A2A Agent 卡片a2aAgentCard/.well-known/agent-card.json 包含 name、version 和 supportedInterfaces
Agent 技能索引agentSkills/.well-known/agent-skills/index.json 包含有效的 skills 数组（也接受旧版 /.well-known/skills/）
WebMCPwebMcp页面通过 navigator.modelContext.provideContext() 暴露 MCP 工具
ARDard/.well-known/ai-catalog.json 包含 specVersion、host 和非空的 entries
Personal Agent Protocol（Poppy）personalAgentProtocol/.well-known/poppy.json（草案 0.1）包含有效的 organization/domain 和接口；如果声明了身份验证，则需有匹配的 OAuth 颁发者，且其元数据在 poppy_domains 中列出该域名（仅供参考——不影响级别或评分）
商务 — 可选（5 项检查，仅在检测到电商信号时计分）
检查项API 密钥通过标准
x402x402API 路由返回 HTTP 402，并包含有效的 x402 支付标头
MPPmpp/openapi.json 中可付费操作包含 x-payment-info 扩展（Machine Payment Protocol）
UCPucp/.well-known/ucp 包含 protocol_version 和 services
ACPacp/.well-known/acp.json 包含 protocol.name、api_base_url、transports、capabilities.services
AP2ap2A2A Agent 卡片包含带角色信息的 AP2 扩展

2. 级别

级别名称要求
0尚未就绪不符合级别 1 的标准
1基本 Web 状态以下 3 项中通过 2 项：robots.txt、sitemap、链接标头
2支持 Bot级别 1 + 两项均满足：AI Bot 规则 AND robots.txt 中的 Content Signals
3Agent 可读级别 2 + Markdown 内容协商
4Agent 集成级别 3 + 以下 4 项中通过 1 项：MCP 服务卡片、A2A Agent 卡片、Agent 技能、API 目录
5Agent 原生级别 4 + 以下 3 项中通过 2 项：Web Bot Auth、所有集成、身份验证元数据（OAuth 或 Auth.md）

3. API 参考

端点
POST https://isitagentready.com/api/scan
Content-Type: application/json
必需标头（Cloudflare 保护）
User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/145.0.0.0 Safari/537.36
Referer: https://isitagentready.com/
Origin: https://isitagentready.com
请求正文
{
  "url": "https://example.com/",
  "enabledChecks": [
    "robotsTxt", "sitemap", "linkHeaders", "dnsAid",
    "markdownNegotiation",
    "robotsTxtAiRules", "contentSignals", "webBotAuth",
    "apiCatalog", "oauthDiscovery", "oauthProtectedResource", "authMd",
    "mcpServerCard", "a2aAgentCard", "agentSkills", "webMcp", "ard",
    "personalAgentProtocol",
    "x402", "mpp", "ucp", "acp", "ap2"
  ]
}

所有检查均为可选项——只运行所需的检查即可。

设置 "format": "agent"，即可获取包含修复说明的 Markdown 响应，而非 JSON（适合直接传递给大模型）。

MCP 服务

扫描器也可作为 MCP 服务使用：

https://isitagentready.com/mcp

使用 url 参数调用 scan_site 工具。请参阅 scan-site/SKILL.md。

已发现的端点（isitagentready.com 本身）

扫描器践行了自己的主张。以下是它自身支持 Agent 的端点：

端点内容
/.well-known/api-catalog包含扫描 API 和 MCP 服务条目的 RFC 9727 链接集
/.well-known/mcp/server-card.jsonMCP Server Card（Streamable HTTP 传输）
/.well-known/mcp.json同一 MCP Server Card（备用路径）
/.well-known/agent-skills/index.json采用 Agent Skills Discovery v0.2.0 格式的 25 项技能
/llms.txt面向大模型的扫描器概览
/llms-full.txt完整文档——涵盖全部 23 项检查、通过标准和级别体系的权威参考
/api/health健康检查（{"status":"ok"}）
/mcp带有 scan_site 工具的 Streamable HTTP MCP 服务

API Catalog 将扫描 API 链接到其文档（/llms-full.txt）、 服务说明（/.well-known/mcp/server-card.json）和健康检查 端点（/api/health）。

响应结构
{
  "url": "https://example.com",
  "scannedAt": "2026-04-18T13:10:58.788Z",
  "level": 1,
  "levelName": "Basic Web Presence",
  "isCommerce": false,
  "commerceSignals": [],
  "nextLevel": {
    "target": 2,
    "name": "Bot-Aware",
    "requirements": [
      {
        "check": "contentSignals",
        "description": "...",
        "shortPrompt": "...",
        "prompt": "Full implementation prompt...",
        "specUrls": ["https://..."],
        "skillUrl": "https://isitagentready.com/.well-known/agent-skills/.../SKILL.md"
      }
    ]
  },
  "checks": {
    "<category>": {
      "<checkKey>": {
        "status": "pass|fail|neutral",
        "message": "Human-readable conclusion",
        "durationMs": 42,
        "evidence": [
          {
            "action": "fetch|parse|conclude",
            "label": "GET /robots.txt",
            "request": { "url": "...", "method": "GET" },
            "response": {
              "status": 200,
              "headers": { "content-type": "..." },
              "bodyPreview": "..."
            },
            "finding": {
              "outcome": "positive|negative|neutral",
              "summary": "..."
            }
          }
        ]
      }
    }
  }
}
状态值
状态图标含义
pass✅检查通过
fail❌需要处理
neutral⬜不适用 / 仅供参考

4. Web 界面与 API 对比

功能APIWeb UI
评分和级别✅✅
检查状态 + 消息✅✅
证据（审计详情）✅ 完整✅ 相同
“如何实现”提示词⚠️ 仅 nextLevel（1 条提示词）✅ 所有未通过的检查
技能 URL⚠️ 仅 nextLevel✅ 所有检查

因此，本技能在下方包含了完整的提示词模板——用于根据 API 数据重建 Web 端质量的提示词。

5. 批量扫描

扫描多个域名时，请在请求之间延迟 2 秒。逐步写入结果，以避免数据丢失：

import json, time, urllib.request

HEADERS = {
    "Content-Type": "application/json",
    "User-Agent": "Mozilla/5.0 ...",
    "Referer": "https://isitagentready.com/",
    "Origin": "https://isitagentready.com"
}
CHECKS = ["robotsTxt","sitemap","linkHeaders","dnsAid","markdownNegotiation",
          "robotsTxtAiRules","contentSignals","webBotAuth","apiCatalog",
          "oauthDiscovery","oauthProtectedResource","authMd","mcpServerCard",
          "a2aAgentCard","agentSkills","webMcp","ard","personalAgentProtocol",
          "x402","mpp","ucp","acp","ap2"]

for domain in domains:
    body = json.dumps({"url": f"https://{domain}/", "enabledChecks": CHECKS}).encode()
    req = urllib.request.Request("https://isitagentready.com/api/scan",
                                data=body, headers=HEADERS)
    with urllib.request.urlopen(req, timeout=90) as resp:
        data = json.loads(resp.read())
    # process data...
    time.sleep(2)

6. 子技能（实现指南）

可发现性
内容
机器人访问控制
API、身份验证、MCP 与技能发现
商务（可选）
Meta
  • scan-site/SKILL.md — 扫描任意网站，检查其是否适合 Agent 使用（包含 MCP 服务端点）

7. 最大化影响的优先顺序

  1. robots.txt + sitemap.xml → 第 1 级
  2. robots.txt 中的 Content Signals → 第 2 级
  3. Markdown for Agents + Link headers → 朝第 3 级迈进
  4. MCP Server Card + Agent Skills Index → 朝第 4 级迈进
  5. OAuth discovery + API Catalog → 实现完整的 Agent 互操作性

8. 各项检查的提示词模板

这些模板复现了 Web UI 中“如何实现 — 粘贴到编码 Agent 中”的提示词。生成报告（步骤 4）时，将 {issue} 占位符替换为 API 响应中的实际 message。

robotsTxt
Goal: Publish /robots.txt with clear crawl rules

Issue: {issue}

Fix: Create /robots.txt at the site root with explicit User-agent directives and allow/disallow rules for key paths. Ensure it is plain text and returns 200.

Skill: https://isitagentready.com/.well-known/agent-skills/robots-txt/SKILL.md

Docs: https://www.rfc-editor.org/rfc/rfc9309

子技能：robots-txt/SKILL.md

sitemap
Goal: Publish /sitemap.xml with canonical URLs

Issue: {issue}

Fix: Generate /sitemap.xml listing canonical URLs, keep it updated on publish, and reference it from /robots.txt.

Skill: https://isitagentready.com/.well-known/agent-skills/sitemap/SKILL.md

Docs: https://www.sitemaps.org/protocol.html

子技能：sitemap/SKILL.md

linkHeaders
Goal: Include Link response headers for agent discovery (RFC 8288)

Issue: {issue}

Fix: Add Link response headers to your homepage that point agents to useful resources. For example: Link: </.well-known/api-catalog>; rel="api-catalog" to advertise your API catalog, or Link: </docs/api>; rel="service-doc" for API documentation. See RFC 8288 for the Link header format and IANA Link Relations for registered relation types.

Skill: https://isitagentready.com/.well-known/agent-skills/link-headers/SKILL.md

Docs: https://www.rfc-editor.org/rfc/rfc8288, https://www.rfc-editor.org/rfc/rfc9727#section-3

子技能：link-headers/SKILL.md

markdownNegotiation
Goal: Return HTML responses as markdown when agents request it

Issue: {issue}

Fix: Enable Markdown for Agents so requests with Accept: text/markdown return a markdown version of your HTML response while HTML stays the default for browsers. Confirm the response uses Content-Type: text/markdown (and x-markdown-tokens if available).

Skill: https://isitagentready.com/.well-known/agent-skills/markdown-negotiation/SKILL.md

Docs: https://developers.cloudflare.com/fundamentals/reference/markdown-for-agents/

子技能：markdown-negotiation/SKILL.md

robotsTxtAiRules
Goal: Add User-agent rules for AI crawlers like GPTBot, Claude-Web, and others

Issue: {issue}

Fix: Add explicit User-agent entries for AI crawlers (GPTBot, OAI-SearchBot, Claude-Web, Google-Extended) with allow/disallow rules that match your policy.

Skill: https://isitagentready.com/.well-known/agent-skills/ai-rules/SKILL.md

Docs: https://www.rfc-editor.org/rfc/rfc9309, https://developers.cloudflare.com/ai-crawl-control/

子技能：ai-rules/SKILL.md

contentSignals
Goal: Declare AI content usage preferences with Content Signals in robots.txt

Issue: {issue}

Fix: Add Content-Signal directives to your robots.txt declaring preferences for ai-train, search, and ai-input. For example:
Content-Signal: ai-train=no, search=yes, ai-input=no

Skill: https://isitagentready.com/.well-known/agent-skills/content-signals/SKILL.md

Docs: https://contentsignals.org/, https://datatracker.ietf.org/doc/draft-romm-aipref-contentsignals/

子技能：content-signals/SKILL.md

webBotAuth
Goal: Let your site identify itself as a bot with Web Bot Auth

Issue: {issue}

Fix: Publish a JWKS at /.well-known/http-message-signatures-directory so your site can identify itself when it sends bot or agent requests. Receiving sites can use it to verify those signed requests.

Skill: https://isitagentready.com/.well-known/agent-skills/web-bot-auth/SKILL.md

Docs: https://datatracker.ietf.org/wg/webbotauth/about/, https://developers.cloudflare.com/bots/reference/bot-verification/web-bot-auth/

子技能：web-bot-auth/SKILL.md

apiCatalog
Goal: Publish an API catalog for automated API discovery (RFC 9727)

Issue: {issue}

Fix: Create /.well-known/api-catalog returning application/linkset+json with a "linkset" array. Each entry should include an "anchor" URL for the API and link relations for service-desc (OpenAPI spec), service-doc (documentation), and status (health endpoint). See RFC 9727 Appendix A for examples.

Skill: https://isitagentready.com/.well-known/agent-skills/api-catalog/SKILL.md

Docs: https://www.rfc-editor.org/rfc/rfc9727, https://www.rfc-editor.org/rfc/rfc9264

子技能：api-catalog/SKILL.md

oauthDiscovery
Goal: Publish OAuth/OIDC discovery metadata so agents can authenticate with your APIs

Issue: {issue}

Fix: If your site has protected APIs, publish /.well-known/openid-configuration (for OpenID Connect) or /.well-known/oauth-authorization-server (for pure OAuth 2.0) with your issuer, authorization_endpoint, token_endpoint, jwks_uri, and grant_types_supported. This allows AI agents to programmatically discover how to authenticate.

Skill: https://isitagentready.com/.well-known/agent-skills/oauth-discovery/SKILL.md

Docs: http://openid.net/specs/openid-connect-discovery-1_0.html, https://www.rfc-editor.org/rfc/rfc8414

子技能：oauth-discovery/SKILL.md

oauthProtectedResource
Goal: Publish OAuth Protected Resource Metadata so agents can discover how to authenticate

Issue: {issue}

Fix: Publish /.well-known/oauth-protected-resource with your resource identifier, authorization_servers (list of OAuth/OIDC issuer URLs that can issue tokens for this resource), and scopes_supported. This tells agents how to obtain access tokens for your protected APIs.

Skill: https://isitagentready.com/.well-known/agent-skills/oauth-protected-resource/SKILL.md

Docs: https://www.rfc-editor.org/rfc/rfc9728

子技能：oauth-protected-resource/SKILL.md

mcpServerCard
Goal: Publish an MCP Server Card for agent discovery

Issue: {issue}

Fix: Serve an MCP Server Card (SEP-1649) at /.well-known/mcp/server-card.json with serverInfo (name, version), transport endpoint, and capabilities. The schema is being standardized at https://github.com/modelcontextprotocol/modelcontextprotocol/pull/2127

Skill: https://isitagentready.com/.well-known/agent-skills/mcp-server-card/SKILL.md

Docs: https://github.com/modelcontextprotocol/modelcontextprotocol/pull/2127

子技能：mcp-server-card/SKILL.md

agentSkills
Goal: Publish an agent skills discovery index

Issue: {issue}

Fix: Publish a skills discovery index at /.well-known/agent-skills/index.json (per the Agent Skills Discovery RFC v0.2.0). Include a $schema field, and a skills array where each entry has name, type, description, url, and a sha256 digest.

Skill: https://isitagentready.com/.well-known/agent-skills/agent-skills/SKILL.md

Docs: https://github.com/cloudflare/agent-skills-discovery-rfc, https://agentskills.io/

子技能：agent-skills/SKILL.md

webMcp
Goal: Support WebMCP to expose site tools to AI agents via the browser

Issue: {issue}

Fix: Implement the WebMCP API by calling navigator.modelContext.provideContext() with tool definitions that expose your site's key actions to AI agents. Each tool needs a name, description, inputSchema (JSON Schema), and an execute callback function.

Skill: https://isitagentready.com/.well-known/agent-skills/webmcp/SKILL.md

Docs: https://webmachinelearning.github.io/webmcp/, https://developer.chrome.com/blog/webmcp-epp

子技能：webmcp/SKILL.md

ard
Goal: Publish an ARD capability manifest at /.well-known/ai-catalog.json

Issue: {issue}

Fix: Serve /.well-known/ai-catalog.json with Content-Type application/json, HTTP 200, and Access-Control-Allow-Origin: *. Include specVersion, a host object (displayName + identifier), and a non-empty entries array. Each entry needs identifier, displayName, type, exactly one of url or data, and 2-5 representativeQueries. Use urn:air:<fqdn>:<namespace>:<name> for entry identifiers.

Skill: https://isitagentready.com/.well-known/agent-skills/ard/SKILL.md

Docs: https://agenticresourcediscovery.org/, https://github.com/Agent-Card/ai-catalog

子技能：ard/SKILL.md

personalAgentProtocol
Goal: Publish Personal Agent Protocol (Poppy) discovery at /.well-known/poppy.json

Issue: {issue}

Fix: Publish the manifest and any required RFC 8414 metadata. Website-only declarations can use web: {} without auth. When agent protocols, APIs, or a browser session endpoint are declared, set auth.issuer and publish issuer metadata with a matching issuer, poppy_domains containing organization.domain, and HTTPS token_endpoint and revocation_endpoint. Informational: does not affect level or score.

Skill: https://isitagentready.com/.well-known/agent-skills/personal-agent-protocol/SKILL.md

Docs: https://personalagentprotocol.org/docs/spec#3-discovery

子技能：personal-agent-protocol/SKILL.md

x402
Goal: Support x402 protocol for agent-native HTTP payments

Issue: {issue}

Fix: Add x402 payment middleware to your API routes to enable AI agents to pay for access via HTTP. Use @x402/express, @x402/hono, or @x402/next middleware with a facilitator URL and wallet address. Protected routes will return HTTP 402 with payment requirements that agents can fulfill automatically.

Skill: https://isitagentready.com/.well-known/agent-skills/x402/SKILL.md

Docs: https://x402.org, https://github.com/coinbase/x402, https://docs.x402.org

子技能：x402/SKILL.md

ucp
Goal: Enable content payments via Universal Commerce Protocol

Issue: {issue}

Fix: Serve /.well-known/ucp with protocol version, services, capabilities, and endpoints, and ensure spec URLs and schemas are reachable.

Skill: https://isitagentready.com/.well-known/agent-skills/ucp/SKILL.md

Docs: https://ucp.dev/specification/overview/

子技能：ucp/SKILL.md

acp
Goal: Publish ACP discovery metadata so agents can discover your commerce API

Issue: {issue}

Fix: Serve /.well-known/acp.json at the origin root with protocol.name "acp", protocol.version, api_base_url, supported transports, and capabilities.services so agents can discover your ACP implementation without creating a checkout session first.

Skill: https://isitagentready.com/.well-known/agent-skills/acp/SKILL.md

Docs: https://agenticcommerce.dev, https://github.com/agentic-commerce-protocol/agentic-commerce-protocol/pull/137

子技能：acp/SKILL.md

a2aAgentCard
Goal: Publish an A2A Agent Card for agent-to-agent discovery

Issue: {issue}

Fix: Serve JSON at /.well-known/agent-card.json with name, version, description, supportedInterfaces (service URL and transport), capabilities, and skills (each with id, name, description). See the A2A Protocol Specification for the full schema.

Skill: https://isitagentready.com/.well-known/agent-skills/a2a-agent-card/SKILL.md

Docs: https://a2a-protocol.org/latest/specification/, https://a2a-protocol.org/latest/topics/agent-discovery/

子技能：a2a-agent-card/SKILL.md

dnsAid
Goal: Publish DNS for AI Discovery (DNS-AID) records for DNS-based agent discovery

Issue: {issue}

Fix: Publish SVCB or HTTPS records under your domain's _agents namespace (e.g. _a2a._agents.example.com or _index._agents.example.com). Use alpn and port SvcParamKeys with mandatory=alpn,port. Use numeric keyNNNNN names for experimental custom parameters until registered. Sign zones with DNSSEC.

Skill: https://isitagentready.com/.well-known/agent-skills/dns-aid/SKILL.md

Docs: https://datatracker.ietf.org/doc/draft-mozleywilliams-dnsop-dnsaid/, https://www.rfc-editor.org/info/rfc9460

子技能：dns-aid/SKILL.md

authMd
Goal: Publish Auth.md agent registration discovery metadata

Issue: {issue}

Fix: Serve /auth.md from the site root as Markdown with an H1 heading containing "auth.md". Include OAuth Protected Resource Metadata at /.well-known/oauth-protected-resource and Authorization Server metadata. Add an agent_auth block with skill, register_uri, and registration methods. If OAuth is not available, keep /auth.md self-contained with audience, registration endpoints, and credential use.

Skill: https://isitagentready.com/.well-known/agent-skills/auth-md/SKILL.md

Docs: https://auth-md.com

子技能：auth-md/SKILL.md

mpp
Goal: Support MPP (Machine Payment Protocol) for agent-native HTTP payments

Issue: {issue}

Fix: Serve /openapi.json at the site root with HTTP 200. Include x-payment-info extensions on payable operations declaring intent (charge or session), method (tempo, stripe, lightning, card), and amount. Optionally include currency, description, and top-level x-service-info with categories.

Skill: https://isitagentready.com/.well-known/agent-skills/mpp/SKILL.md

Docs: https://mpp.dev, https://paymentauth.org/draft-payment-discovery-00.txt

子技能：mpp/SKILL.md

llmsTxt
Goal: Publish an LLM-friendly overview at /llms.txt

Issue: {issue}

Fix: Serve /llms.txt as plain text (UTF-8) with HTTP 200. Start with an H1 title line, include a short summary paragraph, and link to the most important content sections for agents. Optionally link to /llms-full.txt for expanded content.

Skill: https://isitagentready.com/.well-known/agent-skills/llms-txt/SKILL.md

Docs: https://llmstxt.org/

子技能：llms-txt/SKILL.md

llmsFullTxt
Goal: Publish expanded LLM content at /llms-full.txt

Issue: {issue}

Fix: Serve /llms-full.txt as plain text (UTF-8) with HTTP 200. Include structured, detailed content suitable for LLM ingestion covering key topics, APIs, and documentation. Link to it from your /llms.txt file.

Skill: https://isitagentready.com/.well-known/agent-skills/llms-full-txt/SKILL.md

Docs: https://llmstxt.org/

子技能：llms-full-txt/SKILL.md

ap2
Goal: Declare AP2 support in your A2A Agent Card for agent payments

Issue: {issue}

Fix: Add AP2 extension to your A2A Agent Card at /.well-known/agent-card.json with your role (merchant, shopper, etc.) so agents can discover your payment capabilities.

Docs: https://ap2-protocol.org/

评论

支持 Markdown