Readiness

Site / Agent Readiness crawls a project’s site and scores how discoverable and retrievable it is to AI agents and crawlers. It is separate from the brand-visibility product. Prompt runs and the dashboard measure how AI platforms talk about you; Readiness measures how ready your own site is for them. It lives at /dashboard/readiness.

​

What you see

• An overall readiness score out of 100, with a verdict.
• A breakdown across six check categories.
• Every check, marked pass, warn, or fail.
• Run history: every prior run for the project, with any run reopenable in full detail.

A run scores the site across structural checks, grouped into categories, with maturity-weighted aggregation into the overall score.

​

Running a scan

Start a run from the Readiness surface. Set the scope: the number of pages to crawl and an optional path prefix to confine the crawl. Results stream in as the run executes. The dashboard renders a skeleton, reveals each check as it resolves, and shows a live page-allowance counter so you see consumption in flight. Each run is saved to run history when it finishes. Under the hood, each page is scraped with Firecrawl. Lighthouse / PageSpeed is sampled on a few pages per run rather than every page.

​

Check categories

Six top-level categories, each a pass | warn | fail rollup of its checks:

1. Discoverability — how agents and crawlers find and enumerate content.
2. Content Accessibility — whether content is retrievable in agent-friendly form.
3. Bot Access Control — robots and crawl permissions for agents.
4. Protocol Discovery — discoverable protocols and endpoints (MCP, well-known).
5. Agentic Commerce — commerce-for-agents readiness.
6. Agentic Browsing — agent-browsing affordances.

Each category holds subcategories that carry an issuing authority and a maturity badge — for example XML Sitemaps (Sitemaps.org), Web Linking / Link header RFC 8288 (IETF), Markdown Content Negotiation (Cloudflare), llms.txt (Google), Generative Engine Optimization (Aggarwal et al.), AI Search Optimization (Google). Checks are either site-scoped (one result for the whole site) or page-scoped (one result per crawled page).

​

Scope and allowances

Two enforcement levers, both tiered: The first lever is per-run crawl scope — the pages a single run may crawl. The second is a calendar-month pooled page allowance. Effective scope is min(requested, tier per-run cap, remaining allowance). A run that exceeds your remaining balance is clamped, not blocked: it runs against what’s left. The allowance is a ledger that reserves pages when a run starts and refunds the unused portion when it finishes. It resets on the calendar month. Lighthouse is decoupled from crawl size. A generous crawl budget pairs with a small flat Lighthouse sample (1 / 10 / 50), so raising crawl caps never multiplies the slow Lighthouse calls.

​

Free-tier detail

A Free org runs a full scan and sees the overall score plus every check’s collapsed row. It also gets full detail on the first 2 failing-or-warning checks — a real taste. All other expanded detail (page URLs, fix text, recommendations, evidence) is redacted server-side in the API and returned as null. It is never sent to the client, never CSS-hidden. Paid tiers get full detail on every check.

​

Scheduled runs

Schedule recurring runs by tier:

• Free — none. A schedule request returns 402.
• Starter — weekly, monthly.
• Agency — daily, weekly, monthly.

A due schedule fires one run via an hourly cron. The cron claims each due schedule atomically, so no run is duplicated across replicas.

​

PDF export and sharing

Download PDF exports a run as a PDF — plain or branded with your org logo and brand color, plus standards-credibility marks (authority brand logos and spec links). PDF export requires a paid plan; Free sees an upgrade prompt. Public share links require Agency. A share link is a public, revocable URL anyone can open to view a run. Free and Starter see an upgrade prompt.

​

API

These routes back the Readiness surface. They are not part of the REST API reference.

• POST /api/readiness/run — start a run.
• GET /api/readiness/runs — list runs for a project.
• GET /api/readiness/report/[runId]/pdf — download the run’s PDF.
• POST /api/readiness/report/[runId]/share / DELETE …/share — create or revoke a public share link.
• GET|POST /api/readiness/schedule — manage the schedule.
• POST /api/readiness/agent-journey-waitlist — join the Agent Journey waitlist.

​

How to use it

• Read the score and verdict first, then drill into the failing categories.
• Confine the crawl with a path prefix to score one section of a large site without spending the whole allowance.
• On Free, the first two flagged checks are a real taste; upgrade for full detail, PDF export, and share links.
• Schedule a recurring run so the score tracks your site as it changes.