agent-ready-scan-site — Claude Code skill
agent-ready-scan-site scans a website and validates compliance with AI agent standards. Evaluates 23 checks across five categories: discoverability, content accessibility, bot rules, API/auth discovery, and commerce.
Apache-2.0 ★ 141 updated 2026-10-29
npx skills add fabricioctelles/skills --skill agent-ready-scan-site -a claude-codeWhat agent-ready-scan-site does
The skill sends a request to the scan API and returns readiness from 0 to 5, the status of each check, and fix instructions. It detects robots.txt, sitemap, Link headers, Markdown content negotiation, MCP servers, and agent cards. Output formats: JSON for structured processing or Markdown with actionable steps to reach the next readiness level.
Useful for debugging when an agent cannot access a site—the skill identifies which standard is missing and how to add it.
What it helps with
- validate whether an agent can read site content
- identify missing MCP, robots.txt, or OAuth implementations
- get instructions to enable agent access
- compare readiness levels across multiple sites
- debug missing API or contact metadata
Example prompts for the agent
- Check if my site works with agents
- Scan example.com for agent readiness and show me how to improve it
- Compare readiness scores for three services and highlight which needs MCP support
How to install agent-ready-scan-site
1Skills CLI (Vercel)
npx skills add fabricioctelles/skills --skill agent-ready-scan-site -a claude-code2Manually: put the files in place
~/.claude/skills/scan-site/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/scan-site/
1Skills CLI (Vercel)
npx skills add fabricioctelles/skills --skill agent-ready-scan-site -a codex2Manually: put the files in place
.agents/skills/scan-site/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add fabricioctelles/skills --skill agent-ready-scan-site -a cursor2Manually: put the files in place
.cursor/skills/scan-site/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install fabricioctelles/skills skills/agent-ready-cloudflare/scan-site2Manually: put the files in place
.github/skills/scan-site/SKILL.md1Skills CLI (Vercel)
npx skills add fabricioctelles/skills --skill agent-ready-scan-site -a gemini-cli2Platform CLI
gemini skills install https://github.com/fabricioctelles/skills --path skills/agent-ready-cloudflare/scan-siteManually: put the files in place
.devin/skills/scan-site/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about agent-ready-scan-site
How do I install agent-ready-scan-site in Claude Code?
Run npx skills add fabricioctelles/skills --skill scan-site -a claude-code, then open ~/.claude/skills/scan-site/SKILL.md.
What does this skill do?
It checks whether a site supports AI agents by analyzing robots.txt, sitemap, Link headers, Markdown negotiation, MCP servers, and other standards. Returns a readiness level (0–5) with recommendations.
What are the readiness levels?
Level 0 is Not Ready; level 3 means agents can read content; level 4 includes integrations like MCP or Agent Cards; level 5 adds full auth support and all standards.
Source text
SKILL.md · Apache-2.0
---
name: agent-ready-scan-site
description: >
Sub-skill: Scan any URL for agent readiness. Checks 23 standards across 5 categories,
returns readiness level (0-5), status of all checks, and fix instructions.
---
# Skill: Scan Site for Agent Readiness
## What This Skill Does
Scans any website URL and checks whether it implements the standards and protocols that make it accessible to AI agents. Checks 23 standards across 5 categories: discoverability, content accessibility, bot access control, API/auth/MCP discovery, and commerce. Returns a readiness level (0-5), the status of all 23 checks, and fix instructions for any failing checks.
## When to Use It
- Checking whether a site supports agent protocols (MCP, robots.txt, Link headers, etc.)
- Debugging why a site is missing specific agent-readiness checks
- Getting actionable fix instructions to improve a site's agent readiness
- Comparing agent readiness across multiple sites
## How to Call It
Send a POST request to the scan API:
```
POST https://isitagentready.com/api/scan
Content-Type: application/json
{"url": "https://example.com", "format": "agent"}
```
### Parameters
- `url` (required): The URL to scan. Must be a valid HTTP or HTTPS URL.
- `format` (optional): Set to `"agent"` for a markdown response with fix instructions, or `"json"` (default) for structured data.
### Using the MCP Server
Alternatively, connect to the MCP server at:
```
https://isitagentready.com/mcp
```
Call the `scan_site` tool with a `url` parameter.
## How to Interpret Results
### JSON Format
The response includes:
- `level`: Integer 0-5 indicating overall readiness
- `levelName`: Human-readable level name (e.g., "Agent-Readable")
- `checks`: Object with 5 categories, each containing individual check results
- `nextLevel`: What's needed to reach the next level
- Each check has `status` ("pass", "fail", or "unableToCheck") and a `message`
### Agent Format
Returns a markdown document listing:
- The site's current score (e.g., "3/5 Agent-Readable")
- All failing checks with descriptions and fix instructions
- If all checks pass, a confirmation message
### Level Scale
| Level | Name | Key Requirements |
|-------|------|-----------------|
| 0 | Not Ready | Fewer than 2 of: robots.txt, sitemap, Link headers |
| 1 | Basic Web Presence | 2 of 3: robots.txt, sitemap, Link headers |
| 2 | Bot-Aware | Level 1 + both: AI bot rules in robots.txt, Content Signals |
| 3 | Agent-Readable | Level 2 + markdown content negotiation |
| 4 | Agent-Integrated | Level 3 + 1 of 4: MCP Server Card, A2A Agent Card, agent skills, API catalog |
| 5 | Agent-Native | Level 4 + 2 of 3: Web Bot Auth, all integrations, auth metadata (OAuth or Auth.md) |
Personal Agent Protocol (Poppy) draft discovery is included in all/API scans as
`personalAgentProtocol`. It validates public metadata and issuer binding, and is
informational: it does not affect readiness levels or percentage scores.
© fabricioctelles · Apache-2.0 · fabricioctelles/skills