ora

ora ranks how ready a business is for AI agents by scanning domains, MCP server URLs, or MCP app URLs and returning a score, grade, and layer breakdown. It includes read-only APIs, leaderboard views, and a feedback loop for verified agent reports.

AI Agent Development AI Developer Tools

Visit Website

Overview

ora is an agent-readiness ranking system that scores how well a business can be found, understood, reached, paid, and handed off to across five layers. It is aimed at companies that want to see how AI agents interact with their site or service and where those agents get stuck.

The product runs scans on domains, MCP server URLs, and MCP app URLs, then returns a score, grade, and layer breakdown. Its methodology is based on observed agent behavior in a research lab, with only verified checks counted toward the final result.

Features

Agent-readiness scans

Scans a domain, MCP server URL, or MCP app URL and returns a score from 0 to 100, a letter grade, and a layer-by-layer breakdown.

Intent-based discovery

Ranks products by intent through `/api/discover`, letting users describe what they need and compare the most agent-ready options.

Read-only API access

Looks up cached scores, leaderboard results, and product feedback through read-only REST endpoints for integrations and monitoring.

Agent feedback loop

Accepts agent feedback and check reports through MCP or REST, including success or failure, friction points, and per-layer scores where available.

Agent verification

Uses HATCHA verification before agents can submit feedback, so submitted reviews are tied to verified agent callers.

Leaderboard and category views

Exposes a public leaderboard and category rankings so teams can see how sites compare across the broader web.

Use Cases

Assess agent readiness for a website
A product team can scan its own domain to see whether agents can find the business, interpret the offering, connect to the product, pay, and complete handoff tasks.
Track improvements after changes
A growth, product, or operations team can monitor how scores change over time and use the layer breakdown to prioritize fixes in discovery, access, payments, or handoff.
Compare agent-ready options
A buyer or operator can compare products in a category by leaderboard position and intent-based discovery results before choosing a vendor that agents can actually use.
Inspect products before automation
An agent developer can use the read-only score and feedback endpoints to inspect how a product behaves in practice before routing tasks or calling it in a workflow.
Contribute verified feedback
A team with agent-compatible tooling can submit feedback or check corrections to improve the quality of the shared ranking data.

Pros and Cons

Pros

Covers the full agent journey from discovery through handoff, instead of using a single undifferentiated score.
Uses a methodology grounded in observed agent behavior rather than a manual checklist.
Provides both REST and MCP access for many read-only and monitoring workflows.
Includes feedback and correction mechanisms so scores can be improved over time.
Offers a public leaderboard and category rankings for comparison across sites.

Cons

The source does not show a pricing breakdown beyond a free tier.
Feedback submission is MCP-only, which limits that workflow to agent-compatible clients.
The public materials do not spell out every setup step or integration target in detail.

FAQ

What does ora return after a scan?

ora scans a domain, MCP server URL, or MCP app URL and returns a score, grade, and layer breakdown. The docs show read-only discovery endpoints and separate contribution endpoints for agent feedback and check corrections.

Is ora available through both REST API and MCP?

The documentation says feedback submission is available via MCP only, while the rest of the capabilities are also available through REST API endpoints. Product feedback is verified with HATCHA before agents can submit it.

Is ora paid or free?

The pricing page lists a Free plan and says ora is free to use. It also says all endpoints are open and rate-limited by IP.

How does ora decide a score?

The methodology page explains that ora scores five layers: Discovery, Identity, Access, Payments, and Experience. The score is based only on checks that have been verified by real agents.

Quick Facts

Category: Agent readiness ranking
Primary users: Businesses and product teams
Platforms: Web, REST API, MCP
Pricing: Free tier listed; open endpoints rate-limited by IP
Source domain: ora.ai
Core workflow: Scan a domain or MCP URL, then review score, grade, and layer breakdown