Protal/Audit/aimangatranslator.app
Well-preparedScanned Sep 15, 2:38 PM

https://aimangatranslator.app.

After a 14.1-second audit, aimangatranslator.app earned a A — that's 90/100 weighted across 9 categories. We flagged 3 failing checks.

Checks51 completedDuration0m 14sTimeouts0Enginev2.0 · 9 categories
↻ Re-scanNew scan
90/ 100
Grade A
Well-prepared
Passing
37
Warnings
4
Failing
3
Skipped
7
Rules that didn't apply
Opportunities
0
Emerging standards
ScopeThis audit covers aimangatranslator.app only. We don't check the rest of your site automatically.

Nine categories, weighted.

Category 1 (Crawler Policy) and Category 3 (Agent Readiness) carry the most weight. Category 2 (AI Semantic Files) carries a small weight — signals like llms.txt and skill.md count, but less than rules with stronger public evidence.

Priority fixes

Ranked by impact on the overall score. Fix these first.

Critical
R2.4§02 AI Semantic Files

/skill.md exists

No /skill.md found.

High
R9.2§09 Discoverability

Sitemap URLs reachable (sampled)

1/10 sampled sitemap URLs returned 2xx/3xx (10%).

Medium
R3.5§03 Agent Readiness

.well-known/mcp.json exists

No /.well-known/mcp.json or /.well-known/mcp/server.json found.

AI crawler coverage

Which major crawlers you've explicitly addressed in robots.txt.

9 addressed
OpenAI
GPTBot
TrainingAddressed
Anthropic
ClaudeBot
TrainingAddressed
Google
Google-Extended
TrainingAddressed
Apple
Applebot-Extended
TrainingAddressed
Common Crawl
CCBot
TrainingAddressed
ByteDance
Bytespider
TrainingAddressed
Perplexity
PerplexityBot
RetrievalAddressed
OpenAI
OAI-SearchBot
RetrievalAddressed
OpenAI
ChatGPT-User
User-triggeredAddressed

Findings

Every rule we checked, grouped by category.

§01Crawler Policy7 shown
!
R1.1
robots.txt exists
Parseable but with 1 warning(s):
View in methodology
line 91: unknown directive "host"
Warn
!
R1.2
Major training crawlers are named in robots.txt
Explicit policy for 3 training crawler(s) from the current registry: GPTBot, ClaudeBot, Google-Extended.
View in methodology
Not addressed: Applebot-Extended, CCBot, Bytespider, Meta-ExternalAgent, cohere-ai.
Pass threshold: ≥4 explicit (registry size 8).
Warn
✓
R1.3
Major retrieval crawlers are named in robots.txt
Explicit policy for 3 retrieval crawler(s) from the current registry: OAI-SearchBot, ChatGPT-User, PerplexityBot.
View in methodology
Not addressed: Claude-SearchBot, Claude-User, Google-Agent, Meta-ExternalFetcher, MistralAI-User, DuckAssistBot.
Pass threshold: ≥3 explicit (registry size 9).
Pass
✓
R1.4
No contradictory rules
No wildcard-vs-specific contradictions detected.
View in methodology
Pass
✓
R1.5
No accidental site-wide block
No site-wide wildcard block detected.
View in methodology
Pass
✓
R1.6
Crawl-delay reasonable
No Crawl-delay directives — crawlers use their own pacing (recommended).
View in methodology
Pass
✓
R1.7
Sitemap declared in robots.txt
1 Sitemap directive(s); first one (https://aimangatranslator.app/sitemap.xml) returned 200.
View in methodology
Pass
§02AI Semantic Files4 shown
✓
R2.1
/llms.txt exists
/llms.txt found (5049 bytes).
View in methodology
Pass
✓
R2.2
llms.txt format compliant
llms.txt structure matches the llmstxt.org spec.
View in methodology
Pass
✓
R2.3
/llms-full.txt exists (bonus)
/llms-full.txt found (5374 bytes). Useful for content sites that want LLMs to ingest the full corpus in one fetch.
View in methodology
Pass
✕
R2.4
/skill.md exists
No /skill.md found.
View in methodology
skill.md (agentskills.io) describes what AI agents can do with your product. New standard in 2026, low adoption — adding one is a differentiator.
Generate /skill.md with AIAccount required · free, 10 AI-generated fixes / month
Fail
§03Agent Readiness6 shown
✓
R3.1
No bot-fight challenge on initial fetch
Clean response on first fetch — no anti-bot challenge markers in headers or body. ProtalBot reached your origin without mitigation. (HTTP 200)
View in methodology
Pass
·
R3.2
Forms have non-JS submission paths
No <form> elements on this page — nothing to evaluate.
View in methodology
Skipped
✓
R3.3
Clickable elements use real <button> and <a> tags
96% of clickable elements use real <button> or <a href> tags (48 of 50). Agents can identify them directly.
View in methodology
Pass
✓
R3.4
Interactive controls have accessible names
100% of interactive controls have accessible names (48 of 48). Agents can identify what each action will do.
View in methodology
Pass
✕
R3.5
.well-known/mcp.json exists
No /.well-known/mcp.json or /.well-known/mcp/server.json found.
View in methodology
MCP is the cleanest protocol for agents to act on your site directly. Without a discovery file, agent runtimes (Claude Desktop, Cursor, ChatGPT) can't find your MCP server in one fetch.
Hosted MCP Coming Soon
Fail
·
R3.6
MCP endpoint reachable
No MCP endpoint declared — see R3.5.
View in methodology
Hosted MCP Coming Soon
Skipped
§04Schema.org Structured Data8 shown
✓
R4.1
At least one JSON-LD block
4 valid JSON-LD block(s) found.
View in methodology
Pass
✓
R4.2
@type values are valid Schema.org
All 4 @type values are recognized: Organization, WebSite, SoftwareApplication, FAQPage.
View in methodology
Pass
✓
R4.3
Organization schema on homepage
Organization schema present with name, url, logo, sameAs.
View in methodology
Pass
·
R4.4
Page-type schema matches inferred page type
Homepage — Organization schema check is R4.3.
View in methodology
Skipped
✓
R4.5
Schema required fields complete
All 2 entities have their required fields.
View in methodology
Pass
·
R4.6
Schema content consistent with visible content
No high-impact fields (price, headline, name) found to verify.
View in methodology
Skipped
✓
R4.7
FAQPage schema when applicable
FAQPage schema present.
View in methodology
Pass
✓
R4.8
Multiple JSON-LD blocks don't conflict
4 JSON-LD blocks, 4 entities, no conflicts.
View in methodology
Pass
§05Metadata6 shown
✓
R5.1
<title> length is reasonable
<title> is 50 chars: "AI Manga Translator | Translate Manga Pages Online".
View in methodology
Pass
✓
R5.2
<meta description> describes this page
Meta description is 136 chars: "Upload manga images or PDFs — OCR, translate, and keep speech-bubble layout. Layout-aware AI manga t…"
View in methodology
Pass
✓
R5.3
Open Graph required tags
5 of 5 required OG tags present: og:title, og:description, og:image, og:type, og:url.
View in methodology
All present.
Pass
✓
R5.4
og:image is valid
og:image=https://aimangatranslator.app/imgs/og/og-default.jpg
View in methodology
✓ absolute http(s) URL (https://aimangatranslator.app/imgs/og/og-default.jpg)
✓ width ≥ 600 (og:image:width=1200)
✓ height ≥ 315 (og:image:height=630)
Pass
✓
R5.5
Canonical URL
Canonical points to current page: https://aimangatranslator.app/
View in methodology
Pass
✓
R5.6
Page language declared via <html lang>
<html lang="en">.
View in methodology
Pass
§06HTML Semantics6 shown
✓
R6.1
Exactly one <h1>
Single <h1>: "AI Manga Translator Break Language Barriers".
View in methodology
Pass
✓
R6.2
Heading hierarchy doesn't skip levels
34 headings, no level skips.
View in methodology
Pass
✓
R6.3
Primary content marked with <main>
1 <main> tag(s) present.
View in methodology
Pass
✓
R6.4
Page uses multiple semantic HTML tags
4 semantic HTML tag types in use: nav, header, footer, section.
View in methodology
Pass
✓
R6.5
Image alt coverage
16/16 (100%) <img> elements have an alt attribute.
View in methodology
Pass
✓
R6.6
Link text descriptiveness
0/24 sampled links (0%) use non-informative text ("click here", "read more", etc.).
View in methodology
Pass
§07Rendering4 shown
✓
R7.1
Page title in initial HTML
Title "AI Manga Translator | Translate Manga Pages Online" present in initial HTML.
View in methodology
Pass
✓
R7.2
Meta description in initial HTML
Meta description present in initial HTML.
View in methodology
Pass
·
R7.3
Main content text in initial HTML
Rendering check skipped — Playwright failed to launch in this environment.
View in methodology
Skipped
·
R7.4
JSON-LD in initial HTML
Rendering check skipped — Playwright failed to launch in this environment.
View in methodology
Skipped
§08Performance5 shown
!
R8.1
TTFB (Time to First Byte)
TTFB 1323ms — slow. AI crawlers usually allocate seconds-budget per site; this eats into content-fetch time.
View in methodology
Measured from a single origin; geographic latency may differ for your users.
Warn
✓
R8.2
Response compression enabled
Content-Encoding: br.
View in methodology
Pass
✓
R8.3
Modern HTTP version (HTTP/2 or HTTP/3, not HTTP/1.1)
Server advertises HTTP/3 via Alt-Svc.
View in methodology
Pass
✓
R8.4
Cache-Control allows caching
Cache-Control: public, s-maxage=120, stale-while-revalidate=600.
View in methodology
Pass
✓
R8.5
Initial HTML size
Initial HTML 196KB — well within budget.
View in methodology
Pass
§09Discoverability5 shown
✓
R9.1
sitemap valid + non-empty
https://aimangatranslator.app/sitemap.xml: 1600 URLs.
View in methodology
Pass
✕
R9.2
Sitemap URLs reachable (sampled)
1/10 sampled sitemap URLs returned 2xx/3xx (10%).
View in methodology
Failed:
• https://aimangatranslator.app/ja/free-manga/romantic-killer (0)
• https://aimangatranslator.app/zh/free-manga/chopperman-hirofumi-takei (0)
• https://aimangatranslator.app/ko/free-manga/marriage-like-the-moon-waxes-and-wanes-zim-mizukiku (0)
• https://aimangatranslator.app/ja/free-manga/high-school-exciting (0)
• https://aimangatranslator.app/ko/free-manga/sunshine-in-the-miniature-garden (0)
Fail
·
R9.3
RSS/Atom feed (content sites)
Page does not appear to be a content site — feed not expected.
View in methodology
Skipped
!
R9.4
HTTPS forced + HSTS
http:// redirects to https://, but no Strict-Transport-Security header.
View in methodology
Add HSTS so browsers refuse downgrade attacks (e.g. coffee-shop wifi MITM).
Warn
✓
R9.5
www / non-www consolidated
www and non-www variants resolve to the same canonical host (one redirects to the other).
View in methodology
Pass
Measured from us-east-1 (Virginia)