Skip to content

llmse.ai MCP server

ai.llmse/mcpllmse.aistreamable-httpv1.3.12Scanned

Public MCP server for the LLM Search Engine

10 tools: 3 change things and 7 only read.

Tools: 0 delete, 0 irreversible, 3 write, 7 read.
  • Reads private data · 0
  • Reads text others write · 7
  • Can send data out · 10
  • 1 finding

Tools listed without sign-in. Remote posture

With Anlyon

3 tools here change things. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the llmse.ai MCP server can do

The llmse.ai MCP server is published by llmse.ai, a domain the MCP Registry verified. It has 10 tools. It can create or change data (3 tools: classify_url, select_advertiser and audit), but cannot delete data. 7 tools only read.

7 tools bring in text other people can write (analyze_seo, analyze_eeat, analyze_aeo and 4 more); 10 tools can send data out of your system (classify_url, select_advertiser, analyze_seo and 7 more).

The fixed checks matched 1 finding (credentials-mention). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 11, 2026, version 1.3.12.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

10 tools

Changes things 3Claude Code: ask
  • auditPerform comprehensive audit of a website URL.Can send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Perform comprehensive audit of a website URL. Fetches the URL content ONCE and provides a combined report with: - Classification: category, subcategory, language, sentiment, demographics - SEO Analysis: score, grade, issues, recommendations - EEAT Analysis: experience, expertise, authoritativeness, trustworthiness scores - AEO Analysis: AI answer engine optimization score, metrics, issues, signals (includes full Citation Readiness analysis in the nested 'citation' key) - Advertiser Matching: best-fit advertising networks with scores - Similar Sites: competitor/related sites from the same category This is more efficient than calling classify_url, analyze_seo, analyze_eeat, analyze_aeo, select_advertiser, and find_similar_sites separately as it only fetches the page once. Args: url: The website URL to audit (e.g., "https://example.com"). Returns: Comprehensive audit report with: - url: The analyzed URL - classification: Category, subcategory, language, sentiment, demographics - seo: Score, grade, issues, recommendations - eeat: EEAT score, grade, category scores, issues, signals - aeo: AEO score, grade, metrics, issues, signals (includes citation results) - advertisers: Matched advertising networks with scores - similar_sites: Related sites from the same category (up to 10) - cached: Whether result was from cache
  • classify_urlClassify a website URL into category, subcategory, language, and sentiment.Can send data outNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Classify a website URL into category, subcategory, language, and sentiment. Fetches the URL content and uses AI for classification. Results are cached for fast subsequent lookups. Rate limited to 1 request per minute per domain. Args: url: The website URL to classify (e.g., "https://example.com"). Returns: Classification result with: - url: The normalized URL - category: Main category (e.g., "Sports", "Technology") - subcategory: Specific subcategory - language: Detected content language - sentiment: Content sentiment (Good/Neutral/Bad) - age: Target age group (if available) - gender: Target gender (if available) - cached: Whether result was from cache
  • select_advertiserSelect the best advertisers based on website demographics.Can send data outNo findings

    Why it is in this group

    • The name "select_advertiser" says "select": a word for a change.

    What it can reach

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Select the best advertisers based on website demographics. Matches advertisers to website content based on classification demographics. Provide either a URL (classification will be fetched) or demographics directly. Rate limited to 1 request per minute per domain when using URL. Scoring weights: - Category match: +10 points - Age match: +5 points - Gender match: +3 points - Sentiment match: +2 points - Higher CPM bid as tiebreaker Args: url: URL to match advertisers for (fetches classification from cache). category: Target category (e.g., "Sports", "Automotive"). subcategory: Target subcategory. age: Target age group (e.g., "18-24", "25-34", "31-51"). gender: Target gender ("male", "female", or "all"). sentiment: Content sentiment ("Good", "Neutral", or "Bad"). limit: Number of advertisers to return (1-10, default 3). min_cpm: Minimum CPM cost filter (e.g., 5.0 for $5+ CPM). max_cpm: Maximum CPM cost filter (e.g., 10.0 for $10 or less CPM). Returns: Dictionary with: - matches: List of matched advertisers with scores - match_count: Number of matches found - classification: URL classification (if URL provided) - demographics: Provided demographics (if no URL)
Only reads 7Show 7 read-only tools
  • analyze_aeoAnalyze how well content is optimized for AI answer engines.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "analyze_aeo" says "analyze": a word for reading.

    What it can reach

    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Analyze how well content is optimized for AI answer engines. Evaluates content for AI answer engines (ChatGPT, Perplexity, Gemini, Claude). Combines Q&A pattern detection, snippet extractability, and entity clarity analysis with a full Citation Readiness assessment. AEO Scoring Framework (100 points): - Answer Format Detection: 30 points (Q&A extractability patterns) - FAQ Schema Presence: 20 points (FAQPage schema markup) - HowTo Schema Presence: 15 points (HowTo schema markup) - Direct Answer Snippets: 20 points (short extractable blocks <50 words) - Entity Clarity Score: 15 points (clear entity definitions) Neutral Schema Scoring: If no FAQ/HowTo-style content detected, those schema metrics score full points rather than penalizing. Grade Scale: A (85-100), B (70-84), C (55-69), D (40-54), F (0-39) Results are cached for fast subsequent lookups. Rate limited to 1 request per minute per domain. Args: url: The website URL to analyze (e.g., "https://example.com"). Returns: AEO analysis with: - url: The analyzed URL - aeo_score: Overall AEO score (0-100) - aeo_grade: Letter grade (A-F) - aeo_metrics: Individual metric scores - citation: Full Citation Readiness analysis (score, grade, issues, signals) - issues: Problems detected (critical, warnings, info) - signals: Positive signals detected - recommendations: Prioritized improvements - cached: Whether result was from cache
  • analyze_eeatAnalyze a website URL for E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness).Reads text others writeCan send data out1 finding

    Why it is in this group

    • The name "analyze_eeat" says "analyze": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "analyze" and "website": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Findings

    • Info

      The description mentions "credentials".

      Matched text

      description: "...s of experience - Expertise: Author credentials, certifications, professional membershi..."

      What this check looks for
    Description, as the server wrote it
    Analyze a website URL for E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness). Evaluates content quality signals based on Google's Search Quality Rater Guidelines and "Creating helpful content" documentation. Detects EEAT signals including: - Experience: First-person language, case studies, testimonials, years of experience - Expertise: Author credentials, certifications, professional memberships, topic depth - Authoritativeness: Organization schema, awards, trust badges, media mentions - Trustworthiness: HTTPS, contact info, privacy policy, source citations Also detects YMYL (Your Money or Your Life) content for health, financial, and legal topics. Results are cached for fast subsequent lookups. Rate limited to 1 request per minute per domain. Args: url: The website URL to analyze (e.g., "https://example.com"). Returns: EEAT analysis result with: - url: The analyzed URL - score: Overall EEAT score (0-100) - grade: Letter grade (A-F) - scores: Individual category scores (experience, expertise, authoritativeness, trustworthiness) - issues: Categorized issues (critical, warnings, info) - signals: Detected EEAT signals - meta: Extracted meta information - recommendations: Prioritized list of improvements - cached: Whether result was from cache
  • analyze_garmCompute GARM brand safety score for a website or category.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "analyze_garm" says "analyze": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "website": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Compute GARM brand safety score for a website or category. Based on the GARM (Global Alliance for Responsible Media) Brand Suitability Framework. Maps content categories to 11 GARM sensitive content categories with risk levels (Floor, High, Medium, Low). Can either: 1. Provide a URL - classification will be fetched and mapped to GARM 2. Provide category and sentiment directly for instant scoring Score interpretation: higher = safer for advertising. Floor categories (e.g., Adult) always score 0/F regardless of sentiment. Args: category: LLMSE category (e.g., "Adult", "Politics", "Sports"). sentiment: Content sentiment ("Bad", "Neutral", "Good"). url: Optional URL to analyze (fetches classification from cache). Returns: GARM brand safety analysis with: - score: Brand safety score (0-100, higher = safer) - grade: Letter grade (A-F) - garm_category: Matched GARM category name or None - risk_level: "floor"|"high"|"medium"|"low"|"none" - is_floor: True if not suitable for any advertising - issues: Categorized issues {critical, warnings, info} - recommendations: Improvement suggestions
  • analyze_readabilityAnalyze a website URL for content readability using Flesch Reading Ease.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "analyze_readability" says "analyze": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "analyze" and "website": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Analyze a website URL for content readability using Flesch Reading Ease. Extracts plain text from HTML and computes readability metrics including Flesch Reading Ease score, Flesch-Kincaid grade level, reading time, and word/sentence statistics. Grade Scale (web-optimized): - A (60-100): Easy, 6th-8th grade — ideal for web content - B (50-59): Fairly easy, some high school - C (30-49): Standard, college level - D (10-29): Difficult, graduate level - F (0-9): Very difficult, professional/academic Results are cached for fast subsequent lookups. Rate limited to 1 request per minute per domain. Args: url: The website URL to analyze (e.g., "https://example.com"). Returns: Readability analysis with: - url: The analyzed URL - score: Flesch Reading Ease score (0-100, higher = easier) - grade: Letter grade (A-F) - flesch_kincaid_grade_level: US school grade level equivalent - reading_time_minutes: Estimated reading time in minutes - word_count: Total word count - sentence_count: Total sentence count - difficult_words: Count of difficult/uncommon words - cached: Whether result was from cache
  • analyze_seoAnalyze a website URL for SEO optimizations.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "analyze_seo" says "analyze": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "analyze" and "website": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Analyze a website URL for SEO optimizations. Fetches the URL content and analyzes HTML for possible SEO improvements. Results are cached for fast subsequent lookups. Rate limited to 1 request per minute per domain. Args: url: The website URL to analyze (e.g., "https://example.com"). Returns: SEO analysis result with: - url: The analyzed URL - score: Overall SEO score (0-100) - grade: Letter grade (A-F) - issues: List of SEO issues found (critical, warnings, info) - meta: Extracted meta information (title, description, headings, etc.) - recommendations: Prioritized list of improvements - cached: Whether result was from cache
  • analyze_wcagAnalyze a website URL for WCAG 2.1 Level A accessibility issues.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "analyze_wcag" says "analyze": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "analyze" and "website": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Analyze a website URL for WCAG 2.1 Level A accessibility issues. Automated static HTML analysis covering approximately 30-40% of WCAG 2.1 Level A criteria. Checks include: image alt text, form labels, heading hierarchy, page title, html lang, empty links/buttons, ARIA labels, duplicate IDs, skip navigation, table headers, landmarks, viewport zoom, autoplay media, and tabindex ordering. Manual testing is required for full WCAG compliance assessment. Results are cached for fast subsequent lookups. Rate limited to 1 request per minute per domain. Args: url: The website URL to analyze (e.g., "https://example.com"). Returns: WCAG analysis with: - url: The analyzed URL - score: Accessibility score (0-100) - grade: Letter grade (A-F) - issues: Categorized issues (critical, warnings, info) - meta: Extracted accessibility metadata - recommendations: Prioritized improvements - coverage_note: Disclaimer about automated coverage - cached: Whether result was from cache
  • find_similar_sitesFind similar or competitor websites based on classification.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "find_similar_sites" says "find": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "find" and "website": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Claude Code: ask, because it is tagged outbound.

    Description, as the server wrote it
    Find similar or competitor websites based on classification. Takes a URL, classifies it (or uses cached classification), and returns other websites from the same category and subcategory. Useful for competitive analysis and discovering related content. Rate limited to 1 request per minute per domain. Args: url: The website URL to find similar sites for. limit: Maximum number of similar sites to return (1-50, default 10). Returns: Dictionary with: - url: The input URL (normalized) - classification: The URL's category and subcategory - similar_sites: List of similar URLs from the same category - total_in_category: Total sites in this category/subcategory - cached: Whether the classification was from cache

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write7
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out10
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:26 UTC, unchanged since Oct 29, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
3
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 3 tools that change things (3 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://llmse.ai/mcp: tools/list returned 10 tools to an anonymous client, including classify_url (write), select_advertiser (write), audit (write)

    What this check looks for

Posture changes

  1. Oct 29, 2026
    • Reachable: no to yes
    • HTTP status of an anonymous initialize: none to 200
    • tools/list served without credentials: no to yes
    • Tools that change things, listed without credentials: 0 to 3
    • Protocol version: none to 2025-06-18

Findings

Matches of fixed checks, with the text they matched.

1 info
  • The description mentions "credentials".

    Matched text

    description: "...s of experience - Expertise: Author credentials, certifications, professional membershi..."

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. PostureRemote posture changed. Reachable: no -> yes; HTTP status of an anonymous initialize: none -> 200; tools/list served without credentials: no -> yes (and 2 more).View diff
    • Reachable: no to yes
    • HTTP status of an anonymous initialize: none to 200
    • tools/list served without credentials: no to yes
    • Tools that change things, listed without credentials: 0 to 3
    • Protocol version: none to 2025-06-18

Install

Pinned to version 1.3.12. Values in angle brackets are yours to fill in.

claude mcp add --transport http llmse https://llmse.ai/mcp
claude mcp add --transport http llmse https://llmse.ai/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [],
    "ask": [
      "mcp__llmse__classify_url",
      "mcp__llmse__select_advertiser",
      "mcp__llmse__analyze_seo",
      "mcp__llmse__analyze_eeat",
      "mcp__llmse__analyze_aeo",
      "mcp__llmse__analyze_wcag",
      "mcp__llmse__analyze_garm",
      "mcp__llmse__analyze_readability",
      "mcp__llmse__audit",
      "mcp__llmse__find_similar_sites"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as llmse. Review it before you use it.

Capability tags behind 10 rules
  • mcp__llmse__classify_url outbound
  • mcp__llmse__select_advertiser outbound
  • mcp__llmse__analyze_seo untrusted-input, outbound
  • mcp__llmse__analyze_eeat untrusted-input, outbound
  • mcp__llmse__analyze_aeo untrusted-input, outbound
  • mcp__llmse__analyze_wcag untrusted-input, outbound
  • mcp__llmse__analyze_garm untrusted-input, outbound
  • mcp__llmse__analyze_readability untrusted-input, outbound
  • mcp__llmse__audit outbound
  • mcp__llmse__find_similar_sites untrusted-input, outbound

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/ai.llmse/mcp/verify?hash=ebb1c9f133ec113fa355c9a7092d847aa132ce47e009098f297beafa0486e327" | jq -e '.changed == false'

Pinned hash ebb1c9f133ec, from Nov 11, 2026, 19:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/ai.llmse/mcp/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for ai.llmse/mcp
[![Anlyon Radar](https://radar.anlyon.com/badge/ai.llmse/mcp.svg)](https://radar.anlyon.com/servers/ai.llmse/mcp)

Source: MCP Registry. Synced Oct 21, 2026, 17:42 UTC.

llmse.ai MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the llmse.ai MCP server safe?

No scan can promise that. What its definitions show: 10 tools, 3 of which can change, send or delete something, and 1 finding from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the llmse.ai MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the llmse.ai MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the llmse.ai MCP server have?

10 tools: 3 write, 7 read. classify_url, select_advertiser, audit, analyze_seo, analyze_eeat, analyze_aeo, analyze_wcag, analyze_garm and 2 more.

Does the llmse.ai MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the llmse.ai MCP server send data to outside services?

10 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): classify_url, select_advertiser, analyze_seo, analyze_eeat, analyze_aeo, analyze_wcag and 4 more.

Has the llmse.ai MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 11, 2026.