# Ros Thai: Agent-Direct Score B (84.9%)

Site: https://www.rosthai.co.uk/
Report: https://directindex.org/r/rosthai.co.uk
Scanned: 2026-11-14 (Provisional: the text without JavaScript could not be measured on this scan, so those points are left out of the total.)
Ros Thai is mostly ready for agents.
An AI assistant may or may not get through to Ros Thai's website: a plain request was refused and only a full browser got the page. It can read what Ros Thai offers. It finds a page on the site that may take bookings or orders directly; it hasn't been tested yet.

## Direct links
- Order: https://rosthai.co.uk/delivery?scroll-to-menu=true (on the business's own site, found by the scan of 2026-11-14, not yet tested)
- Menu: https://rosthai.co.uk/delivery?scroll-to-menu=true (from the website, 2026-11-14)
- Phone: +44 7463791507 (from the website, 2026-11-14)
- Address: High Street 30 BN2 7HR Brighton (from OpenStreetMap)
- Opening hours: Mon–Tue 17:00–21:30 · Wed–Thu 15:00–21:30 · Fri–Sat 15:00–22:30 · Sun closed (from the website, 2026-11-14)
- Prices: 86 prices on the site, the middle one £10.45
- Is this right? https://directindex.org/about?dispute=rosthai.co.uk#contact

## What assistants answer
The assistants haven't been asked about this business by name yet. They're asked about a sample of each restaurant index and about every claimed business; what they answer is shown here, beside the score, never in it.

Named in none of the 66 answers to 22 everyday questions about Brighton.

Named instead. In the answers that didn't name Ros Thai, the places named most often, and in how many answers.
- Terre à Terre (https://directindex.org/r/terreaterre.co.uk): in 11 answers
- The Salt Room (https://directindex.org/r/saltroom-restaurant.co.uk): in 11 answers
- Burnt Orange (https://directindex.org/r/burnt-orange.co.uk): in 6 answers
- Purezza (https://directindex.org/r/purezza.co.uk): in 6 answers
- Riddle & Finns (https://directindex.org/r/riddleandfinns.co.uk): in 5 answers
See who was named: https://directindex.org/city/brighton/recommended

## What to fix first
- **robots.txt blocks ChatGPT-User, Claude-User, Perplexity-User.** [robots_blocks] (+5 points) These are the assistants' search crawlers and the fetchers that read a page when someone asks. Remove the Disallow lines for them, or add explicit Allow rules. Training crawlers can stay out: that costs nothing here. Training crawlers stay out (GPTBot, ClaudeBot, Google-Extended, anthropic-ai, CCBot, Applebot-Extended, Bytespider, meta-externalagent, cohere-ai): the owner's choice, and it costs nothing in the score. Google uses Google-Extended for Gemini's answers too (grounding in Gemini Apps), so blocking it also keeps the site out of those. Passes when robots.txt lets OAI-SearchBot, Claude-SearchBot, PerplexityBot, ChatGPT-User, Claude-User and Perplexity-User fetch the home page. Guide: https://directindex.org/guides/robots-txt-for-ai-agents
- **Only a full browser got the page; a plain request was refused.** [blocked_for_bots] (not scored, worth checking) A request from a data centre got cloudflare_challenge, and a real browser got the page, so the rest was read on the page as the browser showed it. The assistants' own fetchers come from their companies' addresses, and walls like this often let them through, so this does not show that ChatGPT or Claude are refused. Measured, not scored. If an assistant says it can't open the site, ask the host to let ChatGPT-User, Claude-User and Perplexity-User through. Passes when a plain request without a browser gets the page (not scored). Guide: https://directindex.org/guides/blocking-ai-crawlers-without-knowing

## What the site tells visitors (Measured, not scored)
- **86 prices on the landing page, the middle one £10.45.** In Brighton, the middle price of comparable sites that show prices is £10.95.

## Fix kit: What to paste

Left out until tested: https://rosthai.co.uk/delivery?scroll-to-menu=true looks like a booking or ordering page, but no test has confirmed it yet.

robots.txt. Add these lines to /robots.txt and delete any Disallow rule that names these crawlers.

```
# Let the assistants' search crawlers and the fetchers that read a page for a person in: they send people to you.
# Remove any Disallow lines for these names elsewhere in this file.
# Training crawlers (GPTBot, ClaudeBot, Google-Extended, anthropic-ai, CCBot, Applebot-Extended, Bytespider, meta-externalagent, cohere-ai) stay as you set them: keeping them out costs nothing.
# Your Content-Signal line (ai-train=no) stays as it is.
User-agent: ChatGPT-User
Allow: /

User-agent: Claude-User
Allow: /

User-agent: Perplexity-User
Allow: /
```

## Already right
- **Structured data found (Restaurant).** address, phone, opening hours
- **Menu or price list is an HTML page.** Agents can read the items and prices.

## Where it stands
110 of 532 in Brighton. Ahead of 79% of the other sites scanned in this index.

| What the top 10% of the index does | Top 10% | This site |
|---|---|---|
| Take bookings or orders on their own domain | 100% | yes |
| Publish the menu or price list as HTML | 96% | yes |
| Let the assistants in (robots.txt) | 100% | no |
| Structured data with address, phone and hours | 81% | yes |
| Have an llms.txt | 30% | no |


See what an AI agent reads on this site: https://directindex.org/r/rosthai.co.uk/agent-view

## Businesses like Ros Thai in Brighton

| Group | Sites | Avg score | A or B | This site |
|---|---|---|---|---|
| Businesses of the same kind in Brighton | 532 | 59% | 35% | 84.9% |

With the fix above (robots.txt blocks ChatGPT-User, Claude-User, Perplexity-User), the score would be about 94%. Against today's scores in Brighton, that would rank about 48th of 532 (it is 110th today). The others' scores move too, so this is an estimate.

## The direct record
What agents get for Ros Thai, as JSON: https://directindex.org/api/direct/rosthai.co.uk
- Book: none on record
- Order: https://rosthai.co.uk/delivery?scroll-to-menu=true (own site; found by the scan of 2026-11-14, not yet tested)
- Menu: https://rosthai.co.uk/delivery?scroll-to-menu=true (html_page; from the scan)
- Hours: Mo,Tu 17:00-21:30; We,Th 15:00-21:30; Fr,Sa 15:00-22:30 (closed now) (from rosthai.co.uk)
- Phone: +447463791507
- Stable links that count: https://directindex.org/go/rosthai.co.uk/order, https://directindex.org/go/rosthai.co.uk/menu

## Can an agent transact?
Measured, not scored (not part of the Agent-Direct Score, spec 1.1).
- UCP profile (/.well-known/ucp, Google and Shopify's Universal Commerce Protocol): no
- ACP discovery document (/.well-known/acp.json, OpenAI and Stripe's Agentic Commerce Protocol): no
- MCP server card (/.well-known/mcp/server-card.json): no
- Ordering vendor that takes agent orders (Square, Toast, or a vendor whose terms allow agents): no

Only a real browser got this page (a plain fetch got cloudflare_challenge); Readable and Actionable were measured on the rendered page.

## Score breakdown
| Component | Points | Available |
|---|---|---|
| Reachable | 8 | 13 (the text without JavaScript: pending) |
| Readable | 22 | 25 |
| Actionable | 15 | 15 |

Scored on what the scan measures for every site, the same way for everyone. The action test and what the assistants answer are shown beside the score, never in it (specification 1.1, 18 November 2026).

### Who can read the site, by kind of agent
| Kind of agent | What the scan saw | In the score |
|---|---|---|
| Search crawlers (OAI-SearchBot, Claude-SearchBot, PerplexityBot) | let in by robots.txt | 10 points, the two rows together |
| Fetchers that read a page for a person (ChatGPT-User, Claude-User, Perplexity-User) | blocked by robots.txt: ChatGPT-User, Claude-User, Perplexity-User | 10 points, the two rows together |
| Training crawlers | blocked by robots.txt: GPTBot, ClaudeBot, Google-Extended, anthropic-ai, CCBot, Applebot-Extended, Bytespider, meta-externalagent, cohere-ai | not scored |
| Requests under crawler names from Cloudflare's network | refused: GPTBot, ClaudeBot, PerplexityBot | not scored |
| The scanner's own signed request | refused; only a browser got the page | 2 points: the text without JavaScript |

robots.txt has turned the assistants away since the scan of 2026-11-14.

## Agent readiness (Cloudflare)
Not measured for this scan.
- JSON-LD for the business: yes

Own this business? Claim it, correct what agents are told, recheck a fix, or hand the list to your assistant: https://directindex.org/r/rosthai.co.uk/owner
Methodology: https://directindex.org/methodology · Rescan now: https://directindex.org/scan?url=rosthai.co.uk&force=1
