§5 · COMPARE · 5 TOOLS · SPEC STATUS CHECKED SEPTEMBER 2026
Compare agent readiness tools side by side
Choose two to five tools. Every row comes from the same data as the tool pages.
Agent Readiness Compare editors · Spec status checked September 2026
Answer
Pick the tools you are weighing and read across: which of seven jobs each documents, the scanner score where one applies, published pricing, where it runs and how developers use it. The default view shows ora.ai, Cloudflare and Vercel, the first three tools on our Tools page. These tools mostly do different jobs, so the table does not produce a single winner; only the two readiness scanners are scored.
1.Which tools do you want to compare?
2.How do they compare?
| Attribute | ora.ai | Cloudflare | Vercel |
|---|---|---|---|
| Tool page | Read the ora.ai page | Read the Cloudflare page | Read the Vercel page |
| Primary job | Tests agent readiness: a scored scan of public URLs and recorded agent task runs. | Serves and protects sites at the edge, with a public readiness scanner as a companion. | Verifies bots in application code and hosts MCP servers; offers is-agentic.com with ora.ai. |
| J1 Score readiness against a published checklist | ProvidesScan, 4 layers, 119 checks, grades A+ to F | ProvidesIs It Agent Ready, five categories | Partneris-agentic.com, scans run by ora.ai |
| J2 Watch a real agent attempt a task | ProvidesJourney records each step | Not documented | Not documented |
| J3 Serve content agents can read | Not documented | ProvidesMarkdown for Agents, Pro plan and above | Not documentedfor customer sites |
| J4 Control AI crawler access | Checksrobots.txt and bot access in Scan | ProvidesAI Crawl Control, all plans | ProvidesBotID and bot management |
| J5 Expose actions as tools (MCP, WebMCP) | ChecksWebMCP audit; Scan looks for MCP | Providesremote MCP servers with OAuth | ProvidesMCP servers with mcp-handler |
| J6 Verify which agent is calling | ChecksOAuth and Web Bot Auth in AgentReady | ProvidesWeb Bot Auth verification | Providesbot verification supports Web Bot Auth |
| J7 Take payment from agents or crawlers | Checkspayments layer, 6 checks | PartialPay per crawl, private beta, crawlers only | Not documented |
| Jobs provided | 2 of 7 | 5 of 7 | 3 of 7 |
| Scanner score (0-5) | 4.2 / 5 | 3.3 / 5 | Not scored: is-agentic.com scans are run by ora.ai |
| Standards coverage (20%) | 4.5 / 5AgentReady spec covers robots.txt, llms.txt, sitemaps, JSON-LD, markdown negotiation, MCP, OpenAPI, WebMCP, NLWeb, OAuth 2.0, Web Bot Auth, x402, ACP, UCP and MPP. | 4.5 / 5Checks robots.txt, sitemaps, DNS-AID, markdown negotiation, AI bot rules, Web Bot Auth, MCP, OAuth, Agent Skills, WebMCP, x402, MPP, UCP and ACP. | Not scored |
| Real-agent task testing (20%) | 4.5 / 5Journey records a real agent attempting a chosen intent and shows where it stalled; Deep Scan v2 scores what agents did. | 2 / 5Checks published standards; the public page does not describe running an agent through a task. | Not scored |
| Published methodology (15%) | 4.5 / 5Methodology publishes four layers, points per layer (20/30/40/10), check counts (16/41/56/6) and grade bands A+ to F. | 3 / 5Lists five categories and the standards checked; score bands are not published on the page. | Not scored |
| Fix path on the same platform (15%) | 3.5 / 5Fixes are ranked by impact and the MCP server can rescan after each fix; implementation happens on your own stack. | 4.5 / 5Markdown for Agents, AI Crawl Control, Pay per crawl and Web Bot Auth verification are Cloudflare features on the same network. | Not scored |
| General availability (10%) | 3 / 5Free tier needs no key or sign-up; the home page asks new users to join a waitlist. | 5 / 5Public scan page; enter a URL and scan. | Not scored |
| Developer workflow (20%) | 4.5 / 5npx ax audit CLI, REST API, MCP server and a local WebMCP audit (npx ax webmcp-audit). | 2 / 5No CLI or API for the scanner is documented on the public page. | Not scored |
| Pricing (as published) | Free tier: no API key, sign-up or card. Paid plans and prices: not published. New users are asked to join a waitlist. | Is It Agent Ready: public scan page, no price stated on the page. Markdown for Agents: no cost on Pro, Business and Enterprise plans and SSL for SaaS. AI Crawl Control: all plans. Pay per crawl: private beta. Plan prices: not covered in our review. | Prices for BotID and MCP hosting: not covered in our review. is-agentic.com: public scan page. |
| Where it runs | Hosted scanner for public URLs on any host; the CLI runs with npx; MCP server at https://ora.ai/api/mcp. | At the edge, for sites proxied through Cloudflare; the scanner is a public web page. | In your application on Vercel: BotID checks in your handlers, MCP servers as API routes. |
| Developer access | npx ax audit, REST API, ora.ai MCP server, and a local WebMCP audit with npx ax webmcp-audit. | Markdown for Agents is configured in the dashboard, with configuration rules or by API; no CLI or API is documented for the scanner. | BotID checkBotId() (botid@1.5.0 and above) returns isVerifiedBot, verifiedBotName and verifiedBotCategory; the mcp-handler package deploys MCP servers. |
| Alternatives | ora.ai alternatives | Cloudflare alternatives | Vercel alternatives |
Vercel's J1 status is a partnership: is-agentic.com is built by Vercel with ora.ai, and every scan is run by ora.ai.
Job coverage from public pages (Provides > Partial or Partner > Checks > Pending or Not documented). It is not a quality score.
Scanner score (editorial assessment, 0-5). Compares readiness scanners on what their public pages document. It is not a measure of overall product quality.
Head-to-head pages for your selection:
ora.ai with Vercel: no head-to-head page, because ora.ai tests readiness while these tools implement parts of it. Compare them in the table above.
3.Which head-to-head comparisons are there?
4.Where are the alternatives pages?
Want to weight the scanner criteria yourself? Use the scanner calculator.
Frequently asked questions
How do I compare more than two tools?
Select up to five tools with the buttons above the table. The page address keeps your selection, so you can share the link.
Why is there no overall winner?
These tools do different jobs: one tests, another serves content at the edge, another builds WebMCP tools. Ranking them in one list would mislead, so we score only the two readiness scanners and compare the rest job by job.
Where do the statuses come from?
From each vendor's public pages, reviewed in September 2026, using the six status labels defined on the method page. "Not documented" means we did not find it, not that it does not exist.