agent-ready audits websites for AI agent discoverability and safely fixes structured data, trust pages, llms.txt, crawlability, and Ghost Pro deployment gaps.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-ready
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install agent-ready using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is agent-ready?
agent-ready is an Openclaw Skills workflow for making personal and brand websites easier for AI agents, crawlers, research tools, and browser-based assistants to understand. It audits site identity, trust signals, machine-readable data, content accessibility, crawler reachability, and agent-facing files using a scored Ora report plus targeted validation tools.
The skill coordinates an orchestrator with one or more coding agents while enforcing strict execution boundaries. CMS-level changes such as Ghost pages and content are handled directly, repository changes are handed off through structured specifications, and every improvement is verified live before the workflow continues. This makes Openclaw Skills useful for practical AI SEO improvements without confusing platform limitations with genuine technical problems.
agent-ready Use Cases
- Audit a personal website or creator brand for AI agent discoverability before launching an AI-focused SEO improvement plan.
- Identify missing or thin
/about,/contact, and/privacytrust anchor pages. - Add Person JSON-LD structured data with identity details and verified
sameAsprofile links. - Create and validate an agent-readable
llms.txtfile on a Ghost site. - Investigate whether core page content is available in raw HTML without JavaScript execution.
- Confirm that nonexistent URLs return genuine 404 responses instead of misleading 200 responses.
- Separate actionable website issues from Ghost Pro hosting ceilings, such as managed WAF behavior and markdown content negotiation.
- Track Ora scores, verified improvements, and remaining platform-limited items across sessions.
- Hand off theme, edge-function, robots.txt, redirect, or other git-managed changes to a coding agent through the Openclaw Skills workflow.
How agent-ready Works
- Run the Ora audit at
https://ora.shat the start of every session and keep the scored report in context; previous reports are treated as stale after site changes. - Review Ora findings against the nine-item checklist and the scoring guide to distinguish fixable gaps, false positives, and platform limitations.
- Create or improve trust anchor pages such as
/about,/contact, and/privacythrough the Ghost Admin API, but publish only after the owner approves the exact copy. - Add a Person JSON-LD block through Ghost Pro code injection with the required
@type,name,url,description, andsameAsfields. - Fetch and read existing Ghost head-injection content before appending structured data; never overwrite existing injection code.
- Publish
llms.txtas a Ghost page at/llms.txt/, or use a version-controlled raw file with a redirect when repository ownership is required. - Investigate raw HTML content and 404 behavior manually with curl before making changes, since Ghost server rendering and edge behavior can produce audit false positives.
- Validate completed work using direct URLs, curl, browser snapshots, Google Rich Results Test, Schema.org Validator, llmstxt.cloud, or Lighthouse as appropriate.
- Hand repository-based changes to a coding agent in an
[AGENT_HANDOFF]block instead of editing git-managed files from the orchestrator. - Log the verified Ora score, completed items, and remaining platform-ceiling items in a markdown history table.
agent-ready Setup
Requirements
- An Openclaw Skills-compatible orchestrator with browser access for Ghost Admin tasks.
- Ghost Admin API credentials for pages, posts, images, members, and tags.
- A dedicated Ghost staff account for authenticated browser-session work.
- Access to the site owner for explicit approval of all public-facing copy.
curl,node, andpython3installed locally.- A git-based coding agent such as Codex or Claude Code when repository changes are required.
Installation
Clone the skill repository and enter the skill directory:
git clone https://github.com/highnoonoffice/hno-skills.git
cd hno-skills/agent-ready
Confirm the required command-line tools are available:
curl --version
node --version
python3 --version
Operating procedure
- Open
https://ora.shand run a fresh audit for the target domain. - Read
references/ora-scoring-guide.mdbefore interpreting the score. - Configure the orchestrator with Ghost credentials and browser access, using a dedicated staff account rather than the owner account.
- Prepare exact copy for every new or updated public page and obtain explicit owner approval before publishing.
- For Ghost page updates, fetch the current resource first and include its
updated_atvalue in PUT requests to avoid409conflicts. - Use browser sessions for Ghost Pro code injection, site settings, and redirects; integration tokens alone commonly receive
403responses for these areas. - Verify every completed improvement with a live curl request or browser snapshot before proceeding to the next checklist item.
- Re-run Ora after the session and record the new score and verified changes.
agent-ready Data Schema & Taxonomy
Site artifacts
| Artifact | Location or format | Purpose |
|---|---|---|
| Trust anchor pages | /about, /contact, /privacy |
Establish an identifiable, trustworthy site owner with substantive content of at least 500 characters per page. |
| Person JSON-LD | JSON-LD script in Ghost site header | Provides machine-readable identity, site ownership, description, and external verification links. |
llms.txt |
Ghost page at /llms.txt/, or redirected raw repository file |
Gives AI agents a concise map of who runs the site, what it contains, and where to find key information. |
Optional AI.txt |
/ai.txt via a Ghost page or redirect |
Communicates emerging AI content licensing and permission preferences; distinct from discoverability-focused llms.txt. |
| 404 response | Any nonexistent URL | Must return a genuine HTTP 404 so agents do not mistake missing content for a valid page. |
| Raw HTML content | Initial server response | Used to confirm that core content and an H1 are available without client-side JavaScript. |
JSON-LD taxonomy
@type:Person- Required identity fields:
name,url,description - Verification taxonomy:
sameAsarray containing social profiles, publication bylines, and independently confirming pages - Personal-site guidance: fold organization-level identity into the Person record rather than adding unsupported business fields
- Avoid fabricated
contactPointoraddressvalues when the site does not represent an organization with real business data
Session score log
Maintain one markdown row per audit session:
| Date | Score | Items completed this session | Remaining ceiling items |
|---|---|---|---|
| YYYY-MM-DD | 51 | baseline | all |
| YYYY-MM-DD | 57 | trust pages, JSON-LD, llms.txt | markdown negotiation, WAF |
Record only changes verified as live, and label remaining findings as either actionable or platform-controlled. This history helps Openclaw Skills users determine whether a plateau means all fixable work is complete or an issue was overlooked.
agent-ready Advanced Features
- Mandatory sequential skill gates stop execution when a fresh Ora audit, owner copy approval, pre-injection read, post-change verification, or coding-agent handoff is missing.
- Multi-agent coordination separates CMS and browser responsibilities from repository-based coding work.
- Structured
[AGENT_HANDOFF]blocks support Codex and Claude Code workflows for theme files, edge functions, robots.txt, deployed redirects, and pull-request-based changes. - Ghost Pro-aware deployment logic distinguishes Admin API capabilities from browser-only code injection, settings, and redirect operations.
- Append-only code injection protects existing Ghost header content by requiring a read-before-write step.
- Live verification after every improvement prevents treating an API success response as proof that a change is publicly deployed and correct.
- Tool specialization combines Ora scoring with llmstxt.cloud, Google Rich Results Test, Schema.org Validator, Lighthouse, curl, and browser snapshots.
- False-positive handling prevents unnecessary work on Ghost's server-rendered content, native 404 behavior, unknown crawler signals, and managed WAF limitations.
- Optional
AI.txtsupport helps site owners document AI content rights separately fromllms.txtdiscoverability guidance. - Reference playbooks provide scoring interpretation, Ghost API patterns, JSON-LD templates, llms.txt templates, verification commands, and pull-request review checklists.
SKILL.md
Loading
Related Openclaw Skills
Namecheap DNS Tool
A safety-focused automation skill for managing Namecheap DNS records without accidentally replacing hidden or unmanaged entries.
Agent Telegram
Agent Telegram standardizes how eight specialized AI agents send routed, recognizable progress updates to a Telegram user.
GitHub Trending Feed
GitHub Trending Feed automatically fetches and structures the latest GitHub Trending repositories for developers, teams, and AI-agent workflows.
Token Hub Alternative & AI-HIVE Migration Guide
A practical AI-HIVE migration and Token Hub alternative skill for auditing model compatibility, routing workloads, validating costs, and safely scaling image, video, and commerce generation.
PostKing Reddit Growth
PostKing Reddit Growth finds relevant communities, creates tailored posting angles, and rewrites content into review-ready Reddit drafts.
Alibaba Cloud-Native Internet Diagnostics
A read-only Alibaba Cloud diagnostic skill that verifies public internet egress for cloud-native gateways, SAE applications, and Function Compute functions.
Featured*
