Negotiated 2025-06-18. The bar is 2025-06-18 or newer.
Pass
HTTPS
The endpoint URL is https.
Performance
Not measured
Fast (p50)
Median latency 849 ms over 2 samples in 7 days. The bar is 1000 ms. Need at least 5 samples.
Not measured
Consistent (p95)
95th percentile 918 ms over 2 samples in 7 days. The bar is 3000 ms. Need at least 20 samples.
Not measured
Reliable (uptime)
2 of 2 attempts returned a tool list in 30 days (100%). The bar is 99%. Need at least 30 samples.
Pass
Stable tool list
No tool was removed and no required parameter was added in the last 30 days.
Access
Pass
Open access
tools/list succeeded with no credential, and the record declares none.
Not measured
Standard auth
Auth is not required, so the OAuth metadata check does not apply.
Not measured
Free tier cited
No publisher quote has been checked.
Tools
Pass
Tools described
Every tool description is at least 20 characters.
Pass
Valid input schemas
Every inputSchema is a JSON Schema object.
Pass
Parameters documented
55 of 56 parameters have a description (98%). The bar is 80%.
Pass
Safety hints
Every tool sets readOnlyHint or destructiveHint.
Publisher
Pass
Verified namespace
Listed in the Official MCP Registry.
Pass
Website matches publisher
The website host matches the reverse-domain namespace.
Pass
Source repo linked
A repository URL is on file.
Not measured
Repo owner matches
This namespace is not an io.github login, so the owner comparison does not apply.
Pass
Recently updated
Updated 2026-10-31T03:45:09.574363Z, within 180 days.
Not measured
Open-source licence
The GitHub licence lookup has not returned an OSI result yet.
Start using LeadMarina
Connect to this server
Choose your client, copy the configuration, then enable the server in your client’s MCP settings.
Run this command in your terminal, then restart Claude Code.
Claude Code
claude mcp add --transport http leadmarina https://leadmarina.com/api/mcp
Capabilities
Tools 19
control_advanced_search
action pause or stop takes effect after the batches running now (a queued batch is cancelled and refunded); resume carries on.
Discovered create_automation
Create an automation: re-run a search on a schedule and send the results somewhere automatically. Pass cities as plain strings (e.g. ['Austin, TX', 'Dallas, TX']) — they're resolved for you. IT SPENDS LEADS ON EVERY RUN, one per result, so confirm the niche, cities and cadence with the user before creating one. A destination of 'close', 'ghl' or 'google_sheets' requires that integration to be connected already (Close with no saved field mapping sends Close's own fields); 'csv' | 'xlsx' | 'json' emails a file and needs nothing set up. Every city must resolve and the destination must be one of the listed values, or nothing is created and the error says what to fix. Up to 20 automations, 25 cities each. Automations search US places only for now: a city outside the US is refused. Advanced search's options (a country other than the US — a territory included — max_leads, exact_category) are refused here too.
Discovered delete_automation
Permanently delete an automation. This cannot be undone — confirm with the user first, and prefer update_automation with active=false if they only want it to stop for now. Leads already collected by past runs are kept; only the schedule is removed.
Discovered enrich_domains
Already have businesses? Pass up to 10 website domains to run the full enrichment + verification pipeline on them (verified emails + phones, owner name, socials). Costs 1 lead per distinct domain — spellings of the same site (acme.com, www.acme.com, https://acme.com/contact) count once. Returns a search_id to poll with get_search_results. US businesses only for now: their phones are read as US numbers.
Discovered estimate_advanced_search
What an advanced search (many keywords across many places) would find and how long it would take — without starting it or spending anything. expected_leads is the most likely number of unique leads (each business is charged once however many keywords or places find it); low–high is the range it lands in most of the time.
Discovered export_leads
Push leads into a connected destination: 'close' (CRM — uses the account's saved field mapping), 'ghl' (GoHighLevel sub-account), 'google_sheets' (new spreadsheet, or pass spreadsheet_id for one LeadMarina created before), or a file — 'csv' | 'xlsx' | 'json' (a 7-day download link in the answer; not emailed). Exports EITHER one search's leads (pass search_id) or the whole library (set all=true). Either can be narrowed with the same filter rows as query_leads — with search_id they apply within that search. Leads go out best-ranked first (rank, then id), the order of the leads table. LIMITS PER CALL, which differ by destination because CRMs take one API call per lead: csv/xlsx 50,000 · json 25,000 · google_sheets 5,000 · close/ghl 1,000. For a big library going to a CRM, export csv and import the file instead. If more leads match than the limit, the response sets truncated=true and returns next_cursor — relay that to the user, the export was partial; to send the next batch, call export_leads again with the SAME destination, search_id or all, and filters, plus cursor=next_cursor — for google_sheets also spreadsheet_id from this response, or the next batch goes into a new spreadsheet (up to 50,000 leads a spreadsheet; past that the response has no spreadsheet_id and the next batch starts a new one, or export csv). A next_cursor with not_sent means the destination's time limit cut this batch short: the same call with that cursor sends the rest. An all=true export spread over several calls reads the library live, so a lead that a search running between calls finds again can be skipped or sent twice; for one consistent copy, export the library in a single csv/xlsx (up to 50,000 leads) or json (up to 25,000) call while no search is running. Close, GoHighLevel and a LeadMarina-created sheet passed as spreadsheet_id UPDATE previously exported leads in place (matched by Google CID), never touching the user's own CRM notes/columns — unless update_existing is false (then Close skips leads it already has, and a sheet appends every row); a file or a new sheet is a fresh copy every time — so do not repeat a call that succeeded. Or pass campaign_id instead of search_id to export one advanced search's leads.
Discovered find_places
Where an advanced search can run — the US and its territories for now. With no arguments: the US and its territories (Puerto Rico, Guam …), with their codes. With country 'US': its states and their codes. With q: US places whose name contains q, with their codes and states — for run_advanced_search's places, or search_leads' location_code.
Discovered get_advanced_search
One advanced search: status, leads so far, progress by keyword, its top places and the newest batch running now.
Discovered get_balance
What the account has left. On a plan with a rate limit this is the smaller of what the account holds and what is left in the current 7-day period (period_* describe that week, and period_resets_at is when the week resets); on a metered plan it is the account's lead balance. On a rate-limited plan with a monthly allowance, the note names the allowance when it can be read.
Discovered get_search_results
Fetch a search's leads with verification filled in as enrichment progresses. Poll until enrichment is 'complete'. Each lead includes verified emails (SMTP status), verified phones (line type + carrier), owner name where identified, socials, and the business profile.
Discovered list_advanced_searches
This account's advanced searches, newest first: status, leads so far, searches done of total. The newest `limit`, plus any still running or paused past them.
Discovered list_automations
The account's automations — saved searches that re-run on a schedule and export themselves. Shows each one's niche, cities, cadence, destination, whether it's active, when it next runs, and how the last run went.
Discovered list_integrations
Which export destinations this account has connected (Close, GoHighLevel, Google Sheets) and whether each is ready to receive leads. Check this BEFORE export_leads or create_automation with a CRM destination — those fail if the integration isn't connected. Close with no saved field mapping sends Close's own fields (name, website, address, contact, phone, email). File destinations (csv/xlsx/json) always work and need nothing connected.
Discovered list_searches
The account's recent searches (id, niche, location, result count, when) — use a search_id with get_search_results or export_leads.
Discovered list_views
The account's saved views from the leads table — each a named set of filters the user already built (e.g. 'Roofers with a mobile number'). Returns each view's filter conditions in the exact shape query_leads and export_leads accept, so "export my Hot Leads view to Close" works without rebuilding the filters by hand.
Discovered query_leads
Search ALL leads the account has ever collected (across every search) with filters and sorting — no new search is run and no leads are spent. A Google listing found more than once appears once, with its most recent data. Up to 1,000 leads per call (default 25). When next_cursor is set, call again with cursor = next_cursor for the next page: the cursor carries its filters and sort, so it can be sent alone. Pass search_id to query one search's leads instead of the whole library. Example — plumbers in Austin rated 4.5 or higher with a safe email, most-reviewed first: filters=[{field:'category',operator:'contains',value:'plumb'},{field:'city',operator:'is',value:'Austin',conj:'and'},{field:'rating',operator:'gte',value:'4.5',conj:'and'},{field:'email_status',operator:'is',value:'safe',conj:'and'}], sort=[{id:'reviews',dir:-1}]. A field, operator or value the tool cannot use comes back as an error that names it and lists what works. Pass campaign_id instead of search_id to query one advanced search's leads.
Discovered run_advanced_search
Start an advanced search: many keywords across many US places — the whole US, whole states or single places — run in the background in batches, each business charged once (1 lead), stopping at lead_cap if set. Estimate first with estimate_advanced_search and confirm with the user: it spends leads. Follow it with get_advanced_search; read its leads with query_leads or send them with export_leads (campaign_id). Up to 25 running or paused at once, and 40 new ones (or 250,000 searches) a day.
Discovered search_leads
Run a live search for local businesses by niche + city (e.g. query 'electrician', location 'Austin, TX'). Returns a search_id and, usually, the first batch immediately (its listings are in `results`; their emails and phones are not verified yet — when discovery is 'in_progress' the provider was slow and the whole search was queued, so the leads arrive through get_search_results); more results are found and verified in the background — every email is SMTP-checked and phones are verified with line type + carrier. Spends 1 lead per result. Poll get_search_results until enrichment is 'complete'. If enrichment is already 'complete', nothing was found and nothing more is coming; if capped is true, the balance covered fewer leads than were found. With country, a US territory (PR, GU, VI, AS, MP) is searched; max_leads stops it early and exact_category keeps only that Google category. For many keywords or places at once, use run_advanced_search.
Discovered update_automation
Pause an automation (active=false) so it stops running and stops spending leads, or resume it (active=true). Pausing is the safe way to stop an automation you might want back — use delete_automation only when the user wants it gone for good.