Skip to content

Not Human Search MCP server

ai.nothumansearch/searchnothumansearch.aistreamable-httpv1.7.1Scanned

Search the agentic web. 4,100+ sites, 11 tools incl. check_url + verify_mcp for probe-before-use.

14 tools: 7 change things and 7 only read.

Tools: 0 delete, 0 irreversible, 7 write, 7 read.
  • Reads private data · 0
  • Reads text others write · 2
  • Can send data out · 3
  • No findings

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.
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 Not Human Search MCP server can do

The Not Human Search MCP server is published by nothumansearch.ai, a domain the MCP Registry verified. It has 14 tools. It can create or change data (7 tools: submit_site, check_url, verify_mcp, register_monitor and 3 more), but cannot delete data. 7 tools only read.

2 tools bring in text other people can write (search_agents and submit_site); 3 tools can send data out of your system (submit_site, check_url and verify_mcp).

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

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 11, 2026, version 1.7.1.

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.

14 tools

Changes things 7Claude Code: ask
  • check_urlOn-demand agentic-readiness check for any 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
    On-demand agentic-readiness check for any URL. Runs the NHS 7-signal crawler live (llms.txt, ai-plugin.json, OpenAPI, structured API, MCP server, robots.txt AI rules, Schema.org) and returns a score 0-100 with per-signal breakdown. Use before calling an unfamiliar API to confirm it's agent-usable. Re-runnable without the submissions-table side-effect of submit_site — ideal for verify-before-use workflows.
  • handoff_provider_actionPresent the exact ticket bearer returned by prepare_provider_action.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Present the exact ticket bearer returned by prepare_provider_action. NHS records one privacy-safe handoff receipt, then returns the attributed provider action URL. Optionally and separately, the principal may authorize the exact DNS-verified provider to resolve only the ticket's controlled topic, region, USD budget band, urgency, and allowlisted requirement flags. Declining that optional disclosure does not block the handoff. No query, contact detail, agent/principal identity, network address, referrer, or user-agent is disclosed. The handoff and resolver charge neither party; only the provider's configured downstream outcome can create a provider charge.
  • prepare_provider_actionPrepare a signed action ticket for a separately disclosed provider-funded offer returned beside a receipt-bearing NHS discovery result.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Prepare a signed action ticket for a separately disclosed provider-funded offer returned beside a receipt-bearing NHS discovery result. This step never returns the provider action URL or records a handoff. It requires the discovery receipt and the caller's explicit ticket-authorization attestation. It accepts controlled constraints only—no name, email, contact detail, raw prompt, agent identity, or principal identity fields. Discovery and direct provider access remain free; preparing a ticket charges neither party. Exact v1 wording: https://nothumansearch.ai/privacy#consent-v1
  • record_action_interestRecord the caller's attestation that its human or company principal currently wants one controlled next step with a site returned organically by a receipt-bearing NHS discovery too…No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Record the caller's attestation that its human or company principal currently wants one controlled next step with a site returned organically by a receipt-bearing NHS discovery tool. This creates a private Stage 1 demand receipt that expires with the source discovery receipt, no later than 30 days after it. It does not contact the provider, create an action ticket or charge, change rank or score, or count as commercial proof. No query, contact data, free-form text, or agent/principal identity is accepted. Exact v1 wording: https://nothumansearch.ai/privacy#action-interest-v1
  • register_monitorRegister an email to get alerted when the indicated domain's agentic readiness score drops.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Register an email to get alerted when the indicated domain's agentic readiness score drops. Useful for agents tracking a dependency's agent-readiness health — e.g. an agent that relies on stripe.com's MCP surface wants to know the moment it regresses. Returns an unsubscribe URL. Multiple monitors per email allowed, one per domain.
  • submit_siteSubmit a URL for NHS to crawl and score.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "submit_site" says "submit": a word for a change.

    What it can reach

    • 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
    Submit a URL for NHS to crawl and score. Use when you discover an agent-first tool, API, or service that isn't in the index yet. NHS will fetch the site, check its 7 agentic signals (llms.txt, ai-plugin.json, OpenAPI, structured API, MCP server, robots.txt AI rules, Schema.org), compute a score, and add it to the index. The site becomes searchable within a few seconds if the crawl succeeds.
  • verify_mcpActively probe any URL to check if it is a live, spec-compliant MCP server.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
    Actively probe any URL to check if it is a live, spec-compliant MCP server. Sends a JSON-RPC tools/list request and verifies a valid response. Use this before depending on a third-party MCP endpoint — manifests and documentation can claim MCP support without actually serving it. Returns {verified: true/false, endpoint, note}.
Only reads 7Show 7 read-only tools
  • find_mcp_serversList sites in the index that expose a live MCP server, ranked by agentic readiness.No findings

    Why it is in this group

    • The name "find_mcp_servers" says "find": a word for reading.
    Description, as the server wrote it
    List sites in the index that expose a live MCP server, ranked by agentic readiness. The result set carries a query-free discovery receipt and any eligible provider-funded actions appear separately without changing organic order. Pairs naturally with verify_mcp for a probe-before-use workflow.
  • get_site_detailsGet the full agentic readiness report for a specific domain: score, category, all 7 signal checks (llms.txt, ai-plugin.json, OpenAPI, structured API, MCP server, robots.txt AI rule…No findings

    Why it is in this group

    • The name "get_site_details" says "get": a word for reading.
    Description, as the server wrote it
    Get the full agentic readiness report for a specific domain: score, category, all 7 signal checks (llms.txt, ai-plugin.json, OpenAPI, structured API, MCP server, robots.txt AI rules, Schema.org), plus any cached llms.txt content and OpenAPI summary. When an exact search_id newly records this organic-result selection, the response also carries an optional record_action_interest opportunity for this domain. Selection alone is never interest and never contacts the provider.
  • get_statsCurrent Not Human Search index stats: total sites, average agentic score, top category, sites added in the last 7 days, count of sites exposing an MCP server, and count scoring a p…No findings

    Why it is in this group

    • The name "get_stats" says "get": a word for reading.
    Description, as the server wrote it
    Current Not Human Search index stats: total sites, average agentic score, top category, sites added in the last 7 days, count of sites exposing an MCP server, and count scoring a perfect 100/100.
  • get_top_sitesGet the highest-scored agent-ready sites in the index, optionally filtered by category.No findings

    Why it is in this group

    • The name "get_top_sites" says "get": a word for reading.
    Description, as the server wrote it
    Get the highest-scored agent-ready sites in the index, optionally filtered by category. Organic order is always readiness-score order. A supported category creates a query-free discovery receipt that can support separately disclosed provider-funded actions and explicit action interest; unfiltered browsing stays free and does not enter the Stage 1 demand cohort.
  • list_categoriesList all categories in the Not Human Search index with site counts and average agentic scores.No findings

    Why it is in this group

    • The name "list_categories" says "list": a word for reading.
    Description, as the server wrote it
    List all categories in the Not Human Search index with site counts and average agentic scores. Use this to understand what kinds of agent-ready services exist before searching — counts are live, so the distribution shifts as the index grows.
  • recent_additionsList agent-ready sites newly added to the Not Human Search index, sorted newest first.No findings

    Why it is in this group

    • The description's first verb is "list": a word for reading.
    Description, as the server wrote it
    List agent-ready sites newly added to the Not Human Search index, sorted newest first. A supported category creates a query-free discovery receipt that can support separately disclosed provider-funded actions and explicit action interest; unfiltered catalog watching stays free and does not enter the Stage 1 demand cohort.
  • search_agentsSearch for websites, APIs, and services that AI agents can actually use.Reads text others writeNo findings

    Why it is in this group

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

    What it can reach

    • untrusted-inputThe description's first sentence says "search" and "website": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Search for websites, APIs, and services that AI agents can actually use. Results are ranked by agentic readiness score (0-100) based on llms.txt, OpenAPI specs, ai-plugin.json, structured APIs, and MCP server availability. Use this to discover payment APIs, job boards, data sources, or any web service your agent needs to call.

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 write2
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out3
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 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
2025-06-18
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 7 tools that change things (7 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://nothumansearch.ai/mcp: tools/list returned 14 tools to an anonymous client, including submit_site (write), check_url (write), verify_mcp (write), register_monitor (write), record_action_interest (write) and 2 more

    What this check looks for

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 1.7.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http search https://nothumansearch.ai/mcp
claude mcp add --transport http search https://nothumansearch.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": [
      "mcp__search__search_agents",
      "mcp__search__get_site_details",
      "mcp__search__get_stats",
      "mcp__search__list_categories",
      "mcp__search__get_top_sites",
      "mcp__search__recent_additions",
      "mcp__search__find_mcp_servers"
    ],
    "ask": [
      "mcp__search__submit_site",
      "mcp__search__check_url",
      "mcp__search__verify_mcp",
      "mcp__search__register_monitor",
      "mcp__search__record_action_interest",
      "mcp__search__prepare_provider_action",
      "mcp__search__handoff_provider_action"
    ],
    "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 search. Review it before you use it.

Capability tags behind 4 rules
  • mcp__search__search_agents untrusted-input
  • mcp__search__submit_site untrusted-input, outbound
  • mcp__search__check_url outbound
  • mcp__search__verify_mcp 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.nothumansearch/search/verify?hash=250965414225c0e82dba25a894de584d9704a6ec046a8b2e4e96968a8ffcaa30" | jq -e '.changed == false'

Pinned hash 250965414225, from Nov 11, 2026, 20:18 UTC.

Watch

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

https://radar.anlyon.com/servers/ai.nothumansearch/search/changes.xml

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

Badge

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

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

Not Human Search MCP server: questions and answers

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

Is the Not Human Search MCP server safe?

No scan can promise that. What its definitions show: 14 tools, 7 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 Not Human Search MCP server delete data?

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

Can the Not Human Search 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 Not Human Search MCP server have?

14 tools: 7 write, 7 read. submit_site, check_url, verify_mcp, register_monitor, record_action_interest, prepare_provider_action, handoff_provider_action, search_agents and 6 more.

Does the Not Human Search MCP server need sign-in or OAuth?

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

Does the Not Human Search MCP server send data to outside services?

3 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): submit_site, check_url and verify_mcp.

Has the Not Human Search MCP server changed its tools?

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