sitepulsar.ai MCP server
ai.sitepulsar/mcpsitepulsar.aistreamable-httpv1.0.1Scanned
SitePulsar AEO audits: fetch FIND/READ/USE agent-readiness scores for any website.
13 tools: 7 change things and 6 only read.
- Reads private data · 0
- Reads text others write · 13
- Can send data out · 8
- 2 findings, 1 high
Tools listed without sign-in. Remote posture
With Anlyon
7 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.
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 sitepulsar.ai MCP server can do
The sitepulsar.ai MCP server is published by sitepulsar.ai, a domain the MCP Registry verified. It has 13 tools. It can create or change data (7 tools: check_agent_readiness, run_audit, compare_aeo, probe_agent_discovery and 3 more), but cannot delete data. 6 tools only read.
13 tools bring in text other people can write (check_agent_readiness, run_audit, get_audit and 10 more); 8 tools can send data out of your system (check_agent_readiness, run_audit, compare_aeo and 5 more).
The fixed checks matched 2 findings, 1 of them high severity (annotation-conflict and sensitive-path). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 10, 2026, version 1.0.1.
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.
13 tools
Changes things 7Claude Code: ask
check_agent_readinessFast synchronous AEO / agent-readiness read of a single URL: robots and bot access, structured data (schema), and content structure.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
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 Fast synchronous AEO / agent-readiness read of a single URL: robots and bot access, structured data (schema), and content structure. Returns immediate signals without running a full audit. Use this to triage a page or sanity-check before deciding whether the heavier run_audit is worth a credit.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
compare_aeoCompare AEO posture across multiple URLs (e.g.Reads text others writeCan send data out1 finding
Why it is in this group
- The name "compare_aeo" says "compare": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- 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.
Findings
- Warn
The server marks "compare_aeo" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The name "compare_aeo" says "compare": a word for reading.
Description, as the server wrote it Compare AEO posture across multiple URLs (e.g. a brand versus its competitors) on the same FIND/READ/USE pillar scale. Async: returns an audit_id to poll with get_audit. Spends credits only for freshly-audited URLs; recent audits are reused uncharged.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
probe_agent_discoveryChecks selected registries (official MCP registry, PyPI, GitHub) for packages/servers tied to a domain or brand.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
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 Checks selected registries (official MCP registry, PyPI, GitHub) for packages/servers tied to a domain or brand. A discovery-surface check (can agents find your published tooling?), not a visibility check. Use when you want to know whether a brand has discoverable agent/developer artifacts listed where agents look for them. Result: { state, score, tier, hits[], tool_schema_version }.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
probe_mcp_functionalDiscovers a site's advertised MCP endpoint (mcp.json / .well-known) and inspects its *declared* OAuth/transport posture (advertised, not guaranteed-working — it does not run a full…Reads text others writeCan send data out1 finding
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
- 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.
Findings
- High
The description refers to "mcp.json".
What this check looks forMatched text
description: "...overs a site's advertised MCP endpoint (mcp.json / .well-known) and inspects its *declar..."
Description, as the server wrote it Discovers a site's advertised MCP endpoint (mcp.json / .well-known) and inspects its *declared* OAuth/transport posture (advertised, not guaranteed-working — it does not run a full live handshake). Use when checking whether a site exposes a connectable MCP server and what it claims to support. Result: { handshake_ok, declared_endpoint, declared_tool_names[], score, tool_schema_version }.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
run_auditRun a full AEO audit of a URL covering FIND, READ, and USE.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "run_audit" says "run": 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.
- 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 Run a full AEO audit of a URL covering FIND, READ, and USE. Async: returns an audit_id to poll with get_audit. Accepts an optional target_keyword. Spends one audit credit per fresh run; a same-URL re-run within 24h reuses the cached audit, uncharged.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
scan_product_pageDeterministically scores one product page (schema, price, availability, image) 0–100 for shopping-agent readability — no LLM, fully repeatable.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
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 Deterministically scores one product page (schema, price, availability, image) 0–100 for shopping-agent readability — no LLM, fully repeatable. Use when you want a precise, single-page readability score for a specific product URL rather than a whole-site audit. Available on Pro+ plans. Result: { result: { readability_score, ... }, tool_schema_version }.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
scan_visibilityLive AI-visibility scan for a brand: crawl + reputation sampled across AI engines, returning where *that* brand is mentioned (any public brand, not just your own).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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
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 "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 Live AI-visibility scan for a brand: crawl + reputation sampled across AI engines, returning where *that* brand is mentioned (any public brand, not just your own). Use when you want to know whether and how a named brand already surfaces in AI answers — complementary to search_companies, which finds who agents recommend for a category. Pro+ (LLM cost). Result: { reputation[], tool_schema_version }.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
Only reads 6Show 6 read-only tools
get_auditFetch an audit's status and, when complete, a compact decision-ready SUMMARY: AEO score (overall + FIND/READ/USE), a short summary, the weakest pillar, headline takeaways, and the …Reads text others writeNo findings
Why it is in this group
- The name "get_audit" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Fetch an audit's status and, when complete, a compact decision-ready SUMMARY: AEO score (overall + FIND/READ/USE), a short summary, the weakest pillar, headline takeaways, and the top fixes. Lead with this; call get_audit_detail only when you need the full per-section breakdown.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
get_audit_detailFull structured per-section breakdown of a completed audit, on demand (only call after get_audit when you need depth): per-dimension FIND/READ/USE sub-scores, reputation across AI …Reads text others writeNo findings
Why it is in this group
- The name "get_audit_detail" says "get": 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 "crawl": it reads web pages, which anyone can write.
Description, as the server wrote it Full structured per-section breakdown of a completed audit, on demand (only call after get_audit when you need depth): per-dimension FIND/READ/USE sub-scores, reputation across AI engines, competitor cluster (named for paid tiers), crawl/schema/robots findings, agentic-readiness + USE functional probes, and rendered-DOM analysis (paid). Typed, sanitized, size-capped (see truncated/dropped_sections). Composite *_score fields listed in experimental_fields may change methodology — do not hardcode thresholds. Also surfaces author/E-E-A-T, content freshness, images, hreflang, per-page per-bot access, Schema.org Action microformats, OpenAPI sub-metrics, Google Intelligence, product readability, site maturity, and a methodology block — each tagged with an availability state in the `availability` map (present | not_detected | not_run_free_tier | phase_c_disabled | probe_failed | truncated | not_measured_legacy). Wave C adds deterministic signals: homepage content quality (named quotes, stats-with-source, answer-shape) under crawl.content_signals; per-page video + per-locale schema in page_signals; OpenAPI per-operation coverage %, OAuth scopes, and MCP tool annotations in agentic_detail.use_probes; and self-disclosed trust claims (certifications, SLA/uptime, AI-content disclosure, verifiable-claims) under agentic_detail.trust_claims — each labeled "disclosed"/"mentioned" (never "verified") with an evidence URL and extraction-confidence. All carry an availability state in the `availability` map.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
get_audit_fullOne call that returns a completed audit's SUMMARY, full per-section DETAIL, and deduplicated prioritized FIXES together — so you don't have to chain get_audit → get_audit_detail → …Reads text others writeNo findings
Why it is in this group
- The name "get_audit_full" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it One call that returns a completed audit's SUMMARY, full per-section DETAIL, and deduplicated prioritized FIXES together — so you don't have to chain get_audit → get_audit_detail → get_fixes. Use `expand` to trim the payload ('summary' | 'detail' | 'fixes' | 'all'; default 'all'). Same ownership, tier gating, and sanitization as those tools. For an in-progress audit it returns the status so you can keep polling. The detail layer includes Wave-B surfaced sections (author/E-E-A-T, freshness, images, hreflang, per-bot access, Action microformats, OpenAPI sub-metrics, Google Intelligence, product readability, site maturity, methodology) and Wave-C deterministic signals (content quality, video/locale, USE sub-metrics, trust claims) each with an availability state.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
get_fixesReturn the prioritized, pillar-tagged (FIND / READ / USE) action plan for a completed audit, deduplicated across sources, with machine-actionable implementation steps included on f…Reads text others writeNo findings
Why it is in this group
- The name "get_fixes" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Return the prioritized, pillar-tagged (FIND / READ / USE) action plan for a completed audit, deduplicated across sources, with machine-actionable implementation steps included on fixes where available. Use this when you want the to-do list to act on (or hand to a coding agent), rather than the scores or section detail.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
probe_ucp_readinessInspects /.well-known/ucp to report whether AI shopping agents can transact with the site (presence + advertised capabilities only — never a live purchase).Reads text others writeCan send data outNo findings
Why it is in this group
- The description's first verb is "inspect": 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 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 Inspects /.well-known/ucp to report whether AI shopping agents can transact with the site (presence + advertised capabilities only — never a live purchase). Use when evaluating an e-commerce or merchant site for agentic-commerce readiness. Result: { has_ucp_profile, capabilities[], score, tool_schema_version }.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
search_companiesSamples the major AI engines for which companies they name for a query (e.g.Reads text others writeNo findings
Why it is in this group
- The name "search_companies" says "search": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Samples the major AI engines for which companies they name for a query (e.g. "best CRM for startups"); returns a consensus shortlist (≤5). Use when you want to know who agents *recommend* for a category — not where a specific brand is mentioned (use scan_visibility for that). Free, no URL needed. Result: { companies[], tool_schema_version }.
Annotations the server declares
- readOnlyHint: true
- 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.
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 write13
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out8
- 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.
Looked at Nov 20, 2026, 18:27 UTC, unchanged since Oct 22, 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
- 7
- Protocol version
- 2025-11-25
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://mcp.sitepulsar.ai/.well-known/oauth-protected-resource
- Authorization servers
- https://caring-joy-62.authkit.app
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- yes
- Client ID metadata documents
- yes
- WarnThe whole server
The endpoint served tools/list without credentials, and the list has 7 tools that change things (7 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://mcp.sitepulsar.ai/mcp: tools/list returned 13 tools to an anonymous client, including check_agent_readiness (write), run_audit (write), compare_aeo (write), probe_agent_discovery (write), probe_mcp_functional (write) and 2 more
Findings
Matches of fixed checks, with the text they matched.
The description refers to "mcp.json".
What this check looks forMatched text
description: "...overs a site's advertised MCP endpoint (mcp.json / .well-known) and inspects its *declar..."
- Warncompare_aeo
The server marks "compare_aeo" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The name "compare_aeo" says "compare": a word for reading.
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 1.0.1. Values in angle brackets are yours to fill in.
claude mcp add --transport http sitepulsar https://mcp.sitepulsar.ai/mcpclaude mcp add --transport http sitepulsar https://mcp.sitepulsar.ai/mcpRun 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.
{
"permissions": {
"allow": [
"mcp__sitepulsar__get_audit",
"mcp__sitepulsar__get_audit_detail",
"mcp__sitepulsar__get_fixes",
"mcp__sitepulsar__get_audit_full",
"mcp__sitepulsar__search_companies"
],
"ask": [
"mcp__sitepulsar__check_agent_readiness",
"mcp__sitepulsar__run_audit",
"mcp__sitepulsar__compare_aeo",
"mcp__sitepulsar__probe_agent_discovery",
"mcp__sitepulsar__probe_ucp_readiness",
"mcp__sitepulsar__probe_mcp_functional",
"mcp__sitepulsar__scan_product_page",
"mcp__sitepulsar__scan_visibility"
],
"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 sitepulsar. Review it before you use it.
Capability tags behind 13 rules
- mcp__sitepulsar__check_agent_readiness untrusted-input, outbound
- mcp__sitepulsar__run_audit untrusted-input, outbound
- mcp__sitepulsar__get_audit untrusted-input
- mcp__sitepulsar__get_audit_detail untrusted-input
- mcp__sitepulsar__compare_aeo untrusted-input, outbound
- mcp__sitepulsar__get_fixes untrusted-input
- mcp__sitepulsar__get_audit_full untrusted-input
- mcp__sitepulsar__search_companies untrusted-input
- mcp__sitepulsar__probe_agent_discovery untrusted-input, outbound
- mcp__sitepulsar__probe_ucp_readiness untrusted-input, outbound
- mcp__sitepulsar__probe_mcp_functional untrusted-input, outbound
- mcp__sitepulsar__scan_product_page untrusted-input, outbound
- mcp__sitepulsar__scan_visibility 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.sitepulsar/mcp/verify?hash=cdaa187e4182ce5a2f54a3a728178df80fecb6a8d345e19e7bcda8cbf9eddeee" | jq -e '.changed == false'Pinned hash cdaa187e4182, from Nov 10, 2026, 04:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/ai.sitepulsar/mcp/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/ai.sitepulsar/mcp)Source: MCP Registry · repository. Synced Oct 21, 2026, 17:43 UTC.
sitepulsar.ai MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the sitepulsar.ai MCP server safe?
- No scan can promise that. What its definitions show: 13 tools, 7 of which can change, send or delete something, and 2 findings (1 high) from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the sitepulsar.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 sitepulsar.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 sitepulsar.ai MCP server have?
- 13 tools: 7 write, 6 read. check_agent_readiness, run_audit, compare_aeo, probe_agent_discovery, probe_mcp_functional, scan_product_page, scan_visibility, get_audit and 5 more.
Does the sitepulsar.ai MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the sitepulsar.ai MCP server send data to outside services?
- 8 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): check_agent_readiness, run_audit, compare_aeo, probe_agent_discovery, probe_ucp_readiness, probe_mcp_functional and 2 more.
Has the sitepulsar.ai MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 10, 2026.