Skip to content

Lyre Leads MCP server

com.lyreleads/lyre-leadslyreleads.comstreamable-httpv1.1.0Scanned

Every business of a kind in a city, from Google Maps, read from its website, with verified emails.

16 tools: 7 can delete or cancel things and 9 only read.

Tools: 7 delete, 0 irreversible, 0 write, 9 read.
  • Reads private data · 3
  • Reads text others write · 8
  • Can send data out · 0
  • 3 findings

Tools listed without sign-in. Remote posture

With Anlyon

7 tools here can delete or cancel 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 Lyre Leads MCP server can do

The Lyre Leads MCP server is published by lyreleads.com, a domain the MCP Registry verified. It has 16 tools. It can delete or cancel things (7 tools: search_areas, search_leads, enrich_leads, enroll_leads and 3 more). 9 tools only read.

3 tools read private data (enrich_leads, find_contacts and get_job_status); 8 tools bring in text other people can write (preview_search, search_areas, search_leads and 5 more).

The fixed checks matched 3 findings (annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.

Last scanned Nov 13, 2026, version 1.1.0.

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.

16 tools

Deletes or cancels 7Claude Code: deny
  • enrich_leadsFetch each lead's website and extract 50+ fields: business email, social profiles, CMS, CRM, email-marketing platform, chat widget, booking system, ad pixels, analytics, schema, SS…Reads private dataReads text others writeNo findings

    Why it is in this group

    • The description's first verb is "fetch": a word for reading.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • private-dataThe description's first sentence says "fetch" and "email": it reads mail.
    • 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 "website": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Fetch each lead's website and extract 50+ fields: business email, social profiles, CMS, CRM, email-marketing platform, chat widget, booking system, ad pixels, analytics, schema, SSL, page load, domain age, mail host, and AI-crawler policy. Costs one token per five businesses, on any business you hold, and does not enroll it. The emails it finds are not checked: verify_emails does that, 3 tokens per verified address. To have businesses re-read every month, use enroll_leads. Runs in the background, returns a job id, poll it with get_job_status. Every field follows the three-state contract: a value, a confirmed absence, or unknown.

    Annotations the server declares

    • title: Enrich leads
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • enroll_leadsEnroll leads in automatic enrichment: each is re-read about once a month, free, with change signals when something moves.Reads text others writeNo 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 destructive (destructiveHint: true).

    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
    Enroll leads in automatic enrichment: each is re-read about once a month, free, with change signals when something moves. Nothing is enrolled on its own, so ask the user before enrolling, and offer it after a search for the leads that matched (only_matches, the default). The plan holds up to three months of its leads enrolled; leads past that room are not enrolled and the answer says how many. A lead already read is enrolled for nothing; one never read (an import) is read first at one token per five. Already-enrolled leads and leads without a website are skipped at no cost. When a read is needed it runs in the background and returns a job id: poll get_job_status with kind "enrich".

    Annotations the server declares

    • title: Enroll in enrichment
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • find_contactsTwo paid steps on one job.Reads private dataReads text others write1 finding

    Why it is in this group

    • The name "find_contacts" says "find": a word for reading.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • private-dataThe name "find_contacts" says "find" and "contact": it reads contacts.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Warn

      The server marks "find_contacts" destructive (destructiveHint: true), and its name says "find", a word for reading.

      Matched text

      annotations: {"readOnlyHint":false,"destructiveHint":true,"idempotentHint":false,"openWorldHint":true}; The name "find_contacts" says "find": a word for reading.

      What this check looks for
    Description, as the server wrote it
    Two paid steps on one job. With owner_only true: where a lead names its owner (its own site, or a public registry), guess the owner's own address (at most 3 guesses, no web search) and check it at the mailbox, 10 tokens for each address that comes back verified and nothing otherwise, so the most it can cost is 10 tokens per lead id sent; leads with no owner named are skipped free. Works on every plan, Free included. Without it (paid plans only; a Free account gets a 403 that says so): search the open web for the people behind the business (owner, founder, director), then resolve and verify their addresses, 25 tokens for each contact whose address comes back verified (10 for the owner the lead names) and nothing for a name without one; the run holds 25 tokens per lead id sent and never charges past it. Separate from enrichment, which names the owner where a business's own site does, with no address checked. This is what reaches local business owners who are not in any LinkedIn-derived contact database. Returns a job id, poll with get_job_status. Outcomes per lead: verified (SMTP-confirmed), published (the address is printed on the company's own site but the domain is catch-all so SMTP cannot confirm it), catch_all (person named, no address obtainable), or searched (no name found).

    Annotations the server declares

    • title: Find named contacts
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • search_areasThe search to use for coverage.Reads text others write1 finding

    Why it is in this group

    • The name "search_areas" says "search": a word for reading.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Warn

      The server marks "search_areas" destructive (destructiveHint: true), and its name says "search", a word for reading.

      Matched text

      annotations: {"readOnlyHint":false,"destructiveHint":true,"idempotentHint":false,"openWorldHint":true}; The name "search_areas" says "search": a word for reading.

      What this check looks for
    Description, as the server wrote it
    The search to use for coverage. Sweeps each area with a grid of map points live against Google Maps, as a background job, the same search the Lyre Leads Search page runs. In Houston this found 1,061 roofing contractors where paging from one centre found 255. Costs 1 token per NEW business (one new lead costs one token); listings the account already holds cost nothing (a small per-page minimum applies). Every new business is enriched from its website (the emails it publishes, found and not checked, its owner where the site or a US registry names one) and relevance-checked. Returns job_id: poll get_job_status with kind "search" until finished is true, then read the leads with get_search_results. If an area name fits more than one place, the call returns the candidates with ids and starts nothing: ask the user which one they mean and resend with that id. At most two searches run at once per account.

    Annotations the server declares

    • title: Search whole areas
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • search_leadsSearch ONE city on the user's account and get the businesses back in the same call: 1 to 5 result pages from the city centre, about 20 per page.Reads text others write1 finding

    Why it is in this group

    • The name "search_leads" says "search": a word for reading.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Warn

      The server marks "search_leads" destructive (destructiveHint: true), and its name says "search", a word for reading.

      Matched text

      annotations: {"readOnlyHint":false,"destructiveHint":true,"idempotentHint":false,"openWorldHint":true}; The name "search_leads" says "search": a word for reading.

      What this check looks for
    Description, as the server wrote it
    Search ONE city on the user's account and get the businesses back in the same call: 1 to 5 result pages from the city centre, about 20 per page. Use it when the user asks for a quick search, a page of results or one town; for a whole city or several, use search_areas, which finds several times more. Costs 1 token per NEW business (one new lead costs one token); ones the account already holds are free. Every new business is then enriched from its own website in the background, about two minutes for 20: business email with its verification state, a named contact where the site names one, the software it runs, and a check that it really is the kind of business searched for. Returns the Maps listing and a search_id. Then call list_results with that search_id and wait_seconds 40, and show the user what enrichment found, not only the listing.

    Annotations the server declares

    • title: Quick one-city search
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • unenroll_leadsRemove leads from automatic enrichment.No findings

    Why it is in this group

    • The description's first verb is "remove": a word for removing or ending something.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Remove leads from automatic enrichment. Free and immediate; a removed lead keeps its last reading and its history, and frees room in the plan.

    Annotations the server declares

    • title: Remove from enrichment
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: false
  • verify_emailsSMTP-verify the business emails on saved leads through MillionVerifier.Reads text others writeNo 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 destructive (destructiveHint: true).

    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
    SMTP-verify the business emails on saved leads through MillionVerifier. Costs 3 tokens for each address that comes back verified and nothing for any other answer; an address already verified for that lead within 90 days is not checked or charged again, and the run stops cleanly if the balance runs out. Ask the user first, with the most it can cost (3 tokens per lead with an email), and prefer leads that matched the search. Runs in the background and returns a job id: poll get_job_status with kind "verify" until status is completed, then read the summary. Each address ends in one of four honest states: verified, catch_all (the domain accepts everything, so deliverability cannot be proven), unknown, or invalid. Never treat catch_all as verified.

    Annotations the server declares

    • title: Verify emails
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
Only reads 9Show 9 read-only tools
  • count_saved_leadsFREE and exact.No findings

    Why it is in this group

    • The name "count_saved_leads" says "count": a word for reading.
    Description, as the server wrote it
    FREE and exact. Counts leads THIS ACCOUNT has already searched and saved, matching any combination of filters, with a breakdown of how many are enriched, have an email, have a contact email, or have been relevance-checked. Use it to size a job over data the account already owns before spending: "how many of my saved roofers have no CRM and a verified email". CRITICAL: this is NOT an inventory of what Lyre Leads can find, and a small result here means the account has not searched much yet, nothing more. Never report this number to a user as the amount of data available, and never use it to answer "how many X do you have" — call estimate_list for that.

    Annotations the server declares

    • title: Count saved leads
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • estimate_listFREE, no account needed.No findings

    Why it is in this group

    • The name "estimate_list" says "list": a word for reading.
    Description, as the server wrote it
    FREE, no account needed. Answers "what would it cost to find me N businesses, and what data comes with each one". Returns an exact cost in tokens broken down by stage (search, enrichment, contact discovery; one new lead costs one token), and when the user's account is connected its balance and the maximum businesses affordable right now, and the full list of data points each lead arrives with. THIS is the tool to use when a user asks what Lyre Leads can get them. Do NOT use count_saved_leads for that question: Lyre Leads generates leads on demand from a live Google Maps search and does not sell access to a stored pile of them, so a count of what an account already holds is not a measure of what is available.

    Annotations the server declares

    • title: Price a list
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • find_areasFREE, no account needed.No findings

    Why it is in this group

    • The name "find_areas" says "find": a word for reading.
    Description, as the server wrote it
    FREE, no account needed. Look up the areas search_areas accepts in a country, most populous first, each with an id. Call without q to browse; call with no country to get the list of countries. When a place name is shared (the US list has four Houstons and thirteen Springfields), pass the id of the one the user means to search_areas.

    Annotations the server declares

    • title: Find search areas
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_balanceGet the account's token balance and plan.No findings

    Why it is in this group

    • The name "get_balance" says "get": a word for reading.
    Description, as the server wrote it
    Get the account's token balance and plan. Plans are sold in leads and every price is in tokens: one new lead (a business found by a search, with its website read, the emails it publishes found but not checked, its owner where one is named, and the relevance check) costs one token; a listing (the Google Maps details only) one token per five. Businesses the account already holds: enrichment is one token per five. A verified email costs 3 tokens, charged only when an address comes back verified. The owner's own email costs 10 tokens and a contact found on the open web 25, each charged only when the address comes back verified. Enrolling leads in automatic enrichment (re-read about monthly) is free within the plan's room of three months of its leads. Unused tokens roll over, up to three months' allowance. Plans are bought and changed on the Lyre Leads website, not through these tools; the result carries the link.

    Annotations the server declares

    • title: Token balance
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_job_statusPoll a background job started by search_areas, enrich_leads, enroll_leads (kind "enrich"), find_contacts or verify_emails.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_job_status" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "contact": it reads contacts.
    • untrusted-inputThe description's first sentence says "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Poll a background job started by search_areas, enrich_leads, enroll_leads (kind "enrich"), find_contacts or verify_emails. Pass the job id and its kind. Long jobs are normal: a grid search works through many map points, and enrichment fetches every website in the batch.

    Annotations the server declares

    • title: Check a background job
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_search_resultsFREE.No findings

    Why it is in this group

    • The name "get_search_results" says "get": a word for reading.
    Description, as the server wrote it
    FREE. Read the leads a search_areas job found, up to 100 per page. Accepts the same filters as the Lyre Leads database, passed as filters, e.g. {"relevanceYes": true, "hasVerifiedEmail": true, "noFBPixel": true}. Each lead comes back compact by default: contact details, business email and its verification state, named contact, relevance verdict with its evidence, software and signals, plus a summary and next steps; detail "full" returns every database field. The summary counts every lead in the search (or every lead the filters match), not only the page, so one read answers how many have emails, contacts or no contact yet; read further pages only when the user asks to see more leads. Enrichment keeps landing for a while after the search finishes, so re-read if leads are still enriching.

    Annotations the server declares

    • title: Read a search's leads
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • list_resultsFREE.No findings

    Why it is in this group

    • The name "list_results" says "list": a word for reading.
    Description, as the server wrote it
    FREE. Read saved leads with what enrichment found: business email and its verification state (found, not checked, until verify_emails checks it), the owner where one is named, named contact (name, title, email and its state), the relevance verdict (from the Google category, or with the sentence from the business's own site when the site decided it), software (CMS, CRM, booking, chat, email marketing), technologies and signals. Pass the search_id from search_leads to read that search; wait_seconds (up to 45) first waits for enrichment still running on it. Returns a compact view with a summary and next steps; with a search_id the summary counts the whole search, not only the page. detail "full" returns every field.

    Annotations the server declares

    • title: List saved leads
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false

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 outbound.

Reads private data3
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write8
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0
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, 18:42 UTC, unchanged since Nov 13, 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
https://lyreleads.com/.well-known/oauth-protected-resource/mcp
Authorization servers
https://lyreleads.com
Legacy HTTP+SSE transport
no
Dynamic client registration
yes
Client ID metadata documents
yes
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 7 tools that change things (7 destructive). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://lyreleads.com/mcp: tools/list returned 16 tools to an anonymous client, including search_areas (destructive), search_leads (destructive), enrich_leads (destructive), enroll_leads (destructive), unenroll_leads (destructive) and 2 more

    What this check looks for

Posture changes

  1. Nov 13, 2026
    • Tools that change things, listed without credentials: 6 to 7

Findings

Matches of fixed checks, with the text they matched.

3 warn
  • The server marks "search_areas" destructive (destructiveHint: true), and its name says "search", a word for reading.

    Matched text

    annotations: {"readOnlyHint":false,"destructiveHint":true,"idempotentHint":false,"openWorldHint":true}; The name "search_areas" says "search": a word for reading.

    What this check looks for
  • The server marks "search_leads" destructive (destructiveHint: true), and its name says "search", a word for reading.

    Matched text

    annotations: {"readOnlyHint":false,"destructiveHint":true,"idempotentHint":false,"openWorldHint":true}; The name "search_leads" says "search": a word for reading.

    What this check looks for
  • The server marks "find_contacts" destructive (destructiveHint: true), and its name says "find", a word for reading.

    Matched text

    annotations: {"readOnlyHint":false,"destructiveHint":true,"idempotentHint":false,"openWorldHint":true}; The name "find_contacts" says "find": a word for reading.

    What this check looks for

History

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

  1. Tools1 added, 10 changed.+1 added~10 changedView diff
  2. PostureRemote posture changed. Tools that change things, listed without credentials: 6 -> 7.View diff
    • Tools that change things, listed without credentials: 6 to 7

Install

Pinned to version 1.1.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http lyre-leads https://lyreleads.com/mcp
claude mcp add --transport http lyre-leads https://lyreleads.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__lyre-leads__preview_search",
      "mcp__lyre-leads__find_areas",
      "mcp__lyre-leads__estimate_area_search",
      "mcp__lyre-leads__get_search_results",
      "mcp__lyre-leads__list_results",
      "mcp__lyre-leads__get_balance",
      "mcp__lyre-leads__estimate_list",
      "mcp__lyre-leads__count_saved_leads",
      "mcp__lyre-leads__get_job_status"
    ],
    "ask": [],
    "deny": [
      "mcp__lyre-leads__search_areas",
      "mcp__lyre-leads__search_leads",
      "mcp__lyre-leads__enrich_leads",
      "mcp__lyre-leads__enroll_leads",
      "mcp__lyre-leads__unenroll_leads",
      "mcp__lyre-leads__find_contacts",
      "mcp__lyre-leads__verify_emails"
    ]
  }
}

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 lyre-leads. Review it before you use it.

Capability tags behind 8 rules
  • mcp__lyre-leads__preview_search untrusted-input
  • mcp__lyre-leads__search_areas untrusted-input
  • mcp__lyre-leads__search_leads untrusted-input
  • mcp__lyre-leads__enrich_leads private-data, untrusted-input
  • mcp__lyre-leads__enroll_leads untrusted-input
  • mcp__lyre-leads__find_contacts private-data, untrusted-input
  • mcp__lyre-leads__verify_emails untrusted-input
  • mcp__lyre-leads__get_job_status private-data, untrusted-input

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/com.lyreleads/lyre-leads/verify?hash=1283e6d06f0c494d7f955afbad72887994f3a591cc8654891b30181e9ce68b39" | jq -e '.changed == false'

Pinned hash 1283e6d06f0c, from Nov 13, 2026, 17:17 UTC.

Watch

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

https://radar.anlyon.com/servers/com.lyreleads/lyre-leads/changes.xml

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

Badge

Anlyon Radar badge for com.lyreleads/lyre-leads
[![Anlyon Radar](https://radar.anlyon.com/badge/com.lyreleads/lyre-leads.svg)](https://radar.anlyon.com/servers/com.lyreleads/lyre-leads)

Source: MCP Registry. Synced Oct 21, 2026, 17:58 UTC.

Lyre Leads MCP server: questions and answers

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

Is the Lyre Leads MCP server safe?

No scan can promise that. What its definitions show: 16 tools, 7 of which can change, send or delete something, and 3 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 Lyre Leads MCP server delete data?

Yes. 7 tools can delete, cancel or revoke something: search_areas, search_leads, enrich_leads, enroll_leads, unenroll_leads, find_contacts and 1 more.

Can the Lyre Leads 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 Lyre Leads MCP server have?

16 tools: 7 destructive, 9 read. search_areas, search_leads, enrich_leads, enroll_leads, unenroll_leads, find_contacts, verify_emails, preview_search and 8 more.

Does the Lyre Leads MCP server need sign-in or OAuth?

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

Does the Lyre Leads MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the Lyre Leads MCP server changed its tools?

Yes, once since Anlyon Radar started tracking it, most recently on Nov 13, 2026. Every change is on this page and in its Atom feed.