Skip to content

Black SEO Analyzer MCP server

io.github.sethblack/bseoasethblackstreamable-httpv26.11.18Scanned

Technical SEO audits: check a page or crawl a whole site, locally or hosted. Compares crawls.

11 tools: 3 change things and 8 only read.

Tools: 0 delete, 0 irreversible, 3 write, 8 read.
  • Reads private data · 2
  • Reads text others write · 9
  • Can send data out · 4
  • No findings

Sign-in required, OAuth metadata published. 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 Black SEO Analyzer MCP server can do

The Black SEO Analyzer MCP server is published by the GitHub account sethblack. It has 11 tools. It can create or change data (3 tools: check_page, page_report and start_crawl), but cannot delete data. 8 tools only read.

2 tools read private data (query_issues and search_docs); 9 tools bring in text other people can write (account_status, check_page, compare_crawls and 6 more); 4 tools can send data out of your system (check_page, page_report, serp_preview and 1 more). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.

Its remote endpoint asks for sign-in and publishes OAuth metadata.

Last scanned Nov 17, 2026, version 26.11.18.

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.

11 tools

Changes things 3Claude Code: ask
  • check_pageAudit one URL immediately and return its SEO issues (schema bseoa.issue/v1: key, category, severity, message, fix_hint, docs_url).Reads text others writeCan 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

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • 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
    Audit one URL immediately and return its SEO issues (schema bseoa.issue/v1: key, category, severity, message, fix_hint, docs_url). This is the fast path and the right first call for "what is wrong with this page?" - it fetches a single page and never follows links. Prefer it over start_crawl whenever the user asked about one specific page rather than a whole site, and that includes a site's homepage or bare domain ("check https://example.com/ for SEO problems") when the user did not ask for a crawl, a site audit or a number of pages. Does not count against the account's page budget; limited to 30 calls a minute. JavaScript rendering (spa) is not available on the hosted connector.

    Annotations the server declares

    • title: Check one page
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • page_reportReport everything recorded about one crawled URL: status code, redirect target, title and meta description, Open Graph tags, word count, link and asset counts, headings, TTFB, Web …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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    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
    Report everything recorded about one crawled URL: status code, redirect target, title and meta description, Open Graph tags, word count, link and asset counts, headings, TTFB, Web Vitals score, and the full issue list for that page. Use it after query_issues or issues_summary when you need the context behind a specific page's issues. It never returns the page's HTML source; to inspect markup, fetch the URL yourself.

    Annotations the server declares

    • title: Page report
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • start_crawlQueue a crawl of a whole site and return a job_id immediately; a worker runs the crawl and the pages count against the account's rolling 30-day page budget.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "start_crawl" says "start": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe name "start_crawl" says "crawl": it reads web pages, which anyone can write.
    • untrusted-inputThe description's first sentence says "crawl": 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.
    Description, as the server wrote it
    Queue a crawl of a whole site and return a job_id immediately; a worker runs the crawl and the pages count against the account's rolling 30-day page budget. Use it only when the user asks to crawl or audit a site, or names a number of pages; a request to check a single URL, even a homepage, is check_page, and never start a crawl alongside check_page unasked. Pass the max_pages the user named. One crawl per account runs at a time. Poll crawl_status with the job_id until state is "finished", then read the results with issues_summary and query_issues. max_pages defaults to 500 and is capped at 2000 and at the pages remaining in the budget. JavaScript rendering (spa) is not available on the hosted connector.

    Annotations the server declares

    • title: Start a site crawl
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
Only reads 8Show 8 read-only tools
  • account_statusReport this account's plan and usage live: whether hosted audits are active, the page budget used and remaining in the rolling 30 days and when it resets, how long crawl sessions a…Reads text others writeNo findings

    Why it is in this group

    • The name "account_status" says "status": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "crawl": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Report this account's plan and usage live: whether hosted audits are active, the page budget used and remaining in the rolling 30 days and when it resets, how long crawl sessions are kept, and the licenses on the account. Call it when a tool returned trial_expired, no_hosted_addon or quota_exceeded, when the user says their plan has changed, or when they ask what their plan includes. Read-only: it never changes the account.

    Annotations the server declares

    • title: Account status
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • compare_crawlsCompare two crawls of the same site and report which issues are new in the later crawl, which were fixed, and how many persist.Reads text others writeNo findings

    Why it is in this group

    • The name "compare_crawls" says "compare": a word for reading.

    What it can reach

    • untrusted-inputThe name "compare_crawls" says "crawl": it reads web pages, which anyone can write.
    • untrusted-inputThe description's first sentence says "crawl": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Compare two crawls of the same site and report which issues are new in the later crawl, which were fixed, and how many persist. Use it to answer "did my changes help?" or "what regressed since the last deploy?", and to gate a release: only the new_issues list represents a regression, so a pre-existing backlog never counts against a change. Get both session ids from list_crawls; base_session is the earlier crawl.

    Annotations the server declares

    • title: Compare crawls
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • crawl_statusReport the progress of a crawl started by start_crawl: state ("queued", "running", "finished" or "failed"), the crawl session_id to read results from, the pages stored so far, and …Reads text others writeNo findings

    Why it is in this group

    • The name "crawl_status" says "status": a word for reading.

    What it can reach

    • untrusted-inputThe name "crawl_status" says "crawl": it reads web pages, which anyone can write.
    • untrusted-inputThe description's first sentence says "crawl": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Report the progress of a crawl started by start_crawl: state ("queued", "running", "finished" or "failed"), the crawl session_id to read results from, the pages stored so far, and the error text if it failed. Poll every few seconds; a queued crawl starts as soon as the worker is free. Do not call issues_summary or query_issues for that crawl until state is "finished", or you will read a partial result.

    Annotations the server declares

    • title: Crawl status
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • issues_summarySummarize a crawl: total issue count, counts by severity and by category, and every distinct issue key with how many pages it affects plus up to five example URLs.Reads text others writeNo findings

    Why it is in this group

    • The description's first verb is "summarize": a word for reading.

    What it can reach

    • untrusted-inputThe name "issues_summary" says "issue": it reads issues, which other people can write.
    • untrusted-inputThe description's first sentence says "crawl": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Summarize a crawl: total issue count, counts by severity and by category, and every distinct issue key with how many pages it affects plus up to five example URLs. Always call this before query_issues - it is small, it tells you which keys are worth drilling into, and it is what you should base a prioritized fix plan on. Omit session_id to summarize the account's most recent crawl.

    Annotations the server declares

    • title: Issue summary
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_crawlsList the crawl sessions stored for this account, newest last, with their session_id, start URL, start and end times, and page count.Reads text others writeNo findings

    Why it is in this group

    • The name "list_crawls" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe name "list_crawls" says "crawl": it reads web pages, which anyone can write.
    • untrusted-inputThe description's first sentence says "crawl": it reads web pages, which anyone can write.
    Description, as the server wrote it
    List the crawl sessions stored for this account, newest last, with their session_id, start URL, start and end times, and page count. Single-page checks made with check_page are not crawls and are not listed. Call this first when the user refers to "the last crawl" or when you need a session_id for issues_summary, query_issues, page_report or compare_crawls. A session with no end time is still running or was interrupted.

    Annotations the server declares

    • title: List crawls
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • query_issuesList individual issues from a crawl, filtered by severity, issue key, category or URL prefix, as bseoa.issue/v1 records with the affected URL, the localized message, a fix hint and…Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "query_issues" says "query": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "list" and "message": it reads messages.
    • untrusted-inputThe name "query_issues" says "query" and "issue": it reads issues, which other people can write.
    • untrusted-inputThe description's first sentence says "crawl": it reads web pages, which anyone can write.
    Description, as the server wrote it
    List individual issues from a crawl, filtered by severity, issue key, category or URL prefix, as bseoa.issue/v1 records with the affected URL, the localized message, a fix hint and a docs link. Returns at most 100 issues per call: when the response has a non-null next_offset there are more, so call again with offset set to that value. Narrow with min_severity or key from issues_summary rather than paging through everything.

    Annotations the server declares

    • title: Query issues
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • search_docsSearch Black SEO Analyzer's own documentation and return the matching sections with their URLs.Reads private dataNo findings

    Why it is in this group

    • The name "search_docs" says "search": a word for reading.

    What it can reach

    • private-dataThe name "search_docs" says "search" and "doc": it reads documents.
    Description, as the server wrote it
    Search Black SEO Analyzer's own documentation and return the matching sections with their URLs. Use it for questions about bseoa itself: what a tool, command, flag or setting does, crawl limits and the page budget, licenses and the hosted plan, installing or configuring the CLI, GUI or connector, or what an issue key means (pass issue_key for that issue's reference page with its fix). Pass query for a question, issue_key for one issue, or neither to list the documentation pages. Works whatever the account's hosted status; does not count against the page budget; limited to 30 calls a minute.

    Annotations the server declares

    • title: Search the bseoa documentation
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • serp_previewFetch each URL and report how its title and meta description will look in a search result: character length, pixel width, whether the engine will truncate it, and the Open Graph an…Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "serp_preview" says "preview": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe description's first sentence says "fetch" and "url": it reads web pages, which anyone can write.
    • untrusted-inputThe input takes a URL (input.urls), and the tool reads it: it reads web pages, which anyone can write.
    • outboundThe input takes a URL (input.urls): 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
    Fetch each URL and report how its title and meta description will look in a search result: character length, pixel width, whether the engine will truncate it, and the Open Graph and Twitter Card fallbacks. Use it when the user asks how a page "looks in Google" or is rewriting titles and descriptions - it is about the snippet, not the page's other SEO problems, which check_page covers. At most 50 URLs per call; does not count against the page budget; limited to 30 calls a minute.

    Annotations the server declares

    • title: Search snippet preview
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true

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.

72 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 2 private-data tools breaks every one.

Reads private data2
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write9
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out4
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__bseoa__query_issues",
      "mcp__bseoa__search_docs"
    ]
  }
}

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.

Sign-in required, OAuth metadata published

Looked at Nov 17, 2026, 15:20 UTC, unchanged since Nov 17, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
401
tools/list served without credentials
no
Tools that change things, listed without credentials
0
Protocol version
not observed
WWW-Authenticate on 401
yes
resource_metadata in WWW-Authenticate
https://mcp.blackseoanalyzer.com/.well-known/oauth-protected-resource/mcp
RFC 9728 protected resource metadata
https://mcp.blackseoanalyzer.com/.well-known/oauth-protected-resource/mcp
Authorization servers
https://www.blackseoanalyzer.com
Legacy HTTP+SSE transport
no
Dynamic client registration
yes
Client ID metadata documents
yes

Findings

Matches of fixed checks, with the text they matched.

No check matched anything in these tool definitions. That is not a safety guarantee.

History

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

No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.

Install

Pinned to version 26.11.18. Values in angle brackets are yours to fill in.

claude mcp add --transport http bseoa https://mcp.blackseoanalyzer.com/mcp
claude mcp add --transport http bseoa https://mcp.blackseoanalyzer.com/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": [
      "mcp__bseoa__account_status",
      "mcp__bseoa__compare_crawls",
      "mcp__bseoa__crawl_status",
      "mcp__bseoa__issues_summary",
      "mcp__bseoa__list_crawls",
      "mcp__bseoa__query_issues",
      "mcp__bseoa__search_docs"
    ],
    "ask": [
      "mcp__bseoa__check_page",
      "mcp__bseoa__page_report",
      "mcp__bseoa__serp_preview",
      "mcp__bseoa__start_crawl"
    ],
    "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 bseoa. Review it before you use it.

Capability tags behind 11 rules
  • mcp__bseoa__account_status untrusted-input
  • mcp__bseoa__check_page untrusted-input, outbound
  • mcp__bseoa__compare_crawls untrusted-input
  • mcp__bseoa__crawl_status untrusted-input
  • mcp__bseoa__issues_summary untrusted-input
  • mcp__bseoa__list_crawls untrusted-input
  • mcp__bseoa__page_report outbound
  • mcp__bseoa__query_issues private-data, untrusted-input
  • mcp__bseoa__search_docs private-data
  • mcp__bseoa__serp_preview untrusted-input, outbound
  • mcp__bseoa__start_crawl 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/io.github.sethblack/bseoa/verify?hash=b45e276fb2615a9a12ecf325022342114fa78ca7328605755ba17a74164b9ddc" | jq -e '.changed == false'

Pinned hash b45e276fb261, from Nov 17, 2026, 17:18 UTC.

Watch

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

https://radar.anlyon.com/servers/io.github.sethblack/bseoa/changes.xml

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

Badge

Anlyon Radar badge for io.github.sethblack/bseoa
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.sethblack/bseoa.svg)](https://radar.anlyon.com/servers/io.github.sethblack/bseoa)

Source: MCP Registry · repository. Synced Nov 17, 2026, 17:18 UTC.

Black SEO Analyzer MCP server: questions and answers

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

Is the Black SEO Analyzer MCP server safe?

No scan can promise that. What its definitions show: 11 tools, 3 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Black SEO Analyzer MCP server delete data?

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

Can the Black SEO Analyzer 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 Black SEO Analyzer MCP server have?

11 tools: 3 write, 8 read. check_page, page_report, start_crawl, account_status, compare_crawls, crawl_status, issues_summary, list_crawls and 3 more.

Does the Black SEO Analyzer MCP server need sign-in or OAuth?

Yes. Its remote endpoint answered 401 without credentials, and it publishes OAuth protected resource metadata (RFC 9728); its authorization server supports dynamic client registration.

Does the Black SEO Analyzer MCP server send data to outside services?

4 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): check_page, page_report, serp_preview and start_crawl.

Has the Black SEO Analyzer MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 17, 2026.