MCP App Store
by KeenEthics
Findymail icon

Findymail MCP Server for ChatGPT

By Findymail

Overview

Find verified B2B email addresses, phone numbers and company information with Findymail. Discover employees at a known company, check email deliverability, identify contacts from email addresses, and find similar companies. Describe your ideal customer to start a lead-list search, then retrieve its progress and results. Manage account and team exclusion lists to keep selected companies out of future searches. Requires a Findymail account and authentication. Discovery and enrichment tools consume finder credits; email verification uses a separate verifier-credit pool. Exclusion management and retrieval of existing lead-list results consume no additional credits. This app does not send outreach messages.

How to connect

  1. ChatGPT: Open the Findymail listing in the ChatGPT app directory and choose Connect. ChatGPT listing Findymail exposes 23 tools here.

Tools

add_exclusion_domains

ChatGPT
Add company domains to an exclusion list so they stop appearing in search results. Omit list_id to add them to your global exclusions instead. Domains are reduced to their registrable form before storage: protocol, path and ALL subdomains are stripped, so "careers.acme.com" and "https://www.acme.com/jobs" both become "acme.com" and exclude the whole company. Entries that cannot be parsed as a domain are dropped silently and are not reported back. The response counters do NOT reflect what was stored: total and processed_immediately count the entries you sent, dropped ones included, so total: 4 can mean 3 domains excluded — read the list back with list_exclusion_domains to know what landed. Large batches are accepted (max 10,000): the first 50 domains are applied immediately and the rest are queued in the background, so a search started right away will only account for the first 50. Consumes no credits.

create_contact_list

ChatGPT
Create a new, empty contact list and return it with its id. Fill it by passing that id to find_email (list_id) or generate_lead_list (save_to_list_id), or as lead_list_id to a monitor (create_monitor). Names are not unique: calling this twice creates two lists with the same name, so check list_contact_lists first. The list is private; share it with update_contact_list. Consumes no credits.

create_exclusion_list

ChatGPT
Create a new, empty exclusion list and return its id. Add domains to it with add_exclusion_domains, or let a search fill it automatically via generate_lead_list's add_to_exclusion_list_id. List names must be unique per user. If is_shared is true and you do not belong to a team, the list is still created but the call returns an error — do not retry with the same name, it would then fail as a duplicate. Call list_exclusion_lists to check before retrying. Consumes no credits.

delete_contact_list

ChatGPT
Delete a contact list. This is irreversible: the list and its grouping of contacts are gone for good, for you and for every teammate it was shared with, and there is no undo. The contacts themselves are not deleted — they stay in your saved contacts (list_contacts with list_id 0). Confirm with the user before calling it. Consumes no credits.

delete_monitor

ChatGPT
Delete a signal monitor: it stops watching and stops charging credits at once. This is irreversible — neither the API nor these tools can restore it, for you or for teammates it was shared with; recreating it starts a new monitor from zero. Signals are only readable through a monitor that still exists, so the ones only this monitor caught disappear from list_signals and get_signal with it — export what the user wants to keep first. Confirm with the user before calling it. Consumes no credits.

enrich_company

ChatGPT
Look up firmographic data for one company. Identify it by domain, by LinkedIn company page URL, or by name — at least one is required, and the domain is used first when several are given, since company names are ambiguous. Returns the company profile: name, domain, description, industry, employee-count range, location and social profiles. Use it to qualify a single company before prospecting it, or to fill in a company whose domain came out of find_employees or reverse_email_lookup. To find many companies at once, use lookalike_search or generate_lead_list instead. Costs 1 finder credit per company returned.

find_email

ChatGPT
Find a verified B2B email address for one person. Identify them either by full name plus company domain, or by LinkedIn profile URL — one of those two combinations is required, and the LinkedIn URL takes precedence when both are supplied. Returns the contact with the email address, the person's name and the company domain. Addresses are verified before being returned, so a person whose address cannot be verified comes back as not found rather than as a guess. Use it when you need one specific person's email, and chain it after find_employees (passing the returned name + domain, or the returned LinkedIn URL) to put emails on a list of contacts. Costs 1 finder credit per email found; a lookup that finds nothing is not charged.

find_employees

ChatGPT
Find employees at a company by job title. Given a company website or domain and a list of target job titles, returns matching contacts with name, job title, company and LinkedIn URL. Email addresses are not part of the response: to get a verified email for a contact, pass the returned name and domain — or the returned LinkedIn URL — to find_email. Use it to build a shortlist of people to reach at a known company; to discover the companies themselves, use generate_lead_list or lookalike_search first. Costs 1 finder credit per contact returned.

find_phone

ChatGPT
Find the phone number(s) of a contact, identified by their LinkedIn profile URL. Use it once you already have a LinkedIn URL for the person — returned by find_employees or reverse_email_lookup, or supplied by the user — and need to reach them by phone; a phone number cannot be looked up from a name or an email address. Returns the contact with the mobile and direct-dial numbers found, or an empty result when no number is available. Costs 10 finder credits per phone number found; a lookup that finds nothing is not charged.

generate_lead_list

ChatGPT
Start a company discovery run from a natural language description of an ideal customer profile. This is a semantic search, not a database lookup, so describe the kind of company you are looking for rather than listing exact filter values (e.g. "SaaS companies in the USA with 50+ employees", "tech startups offering insurance services in NYC"). Use 'targeted' mode to return only companies strictly matching the description (narrower) or broad mode to return companies loosely matching the description (broader). Matching companies can be enriched with contacts, verified email addresses and phone numbers. The run is asynchronous and this call returns only a hash: pass that hash to get_lead_list_results to follow progress and retrieve the results. Credit cost depends on the options chosen — 1 finder credit per 10 companies in 'broad' mode or 1 per company in 'targeted' mode, plus 1 per contact found, 1 per email found and 10 per phone number found.

get_company_technologies

ChatGPT
Return the technology stack detected for one company, identified by its domain: each technology with its name, category, subcategory and when it was last detected. Set technologies to only check for specific ones (case-insensitive names, e.g. ["Salesforce", "HubSpot"]) — an empty technologies array in the response then means none of them was detected. A large company can run hundreds of technologies, so at most limit of them are returned (default 50); total gives the full count and truncated is true when some were left out — narrow with technologies or categories rather than raising limit. Use it to qualify an account on its stack before prospecting it. Costs 1 finder credit when the company has at least one technology matching technologies (whatever categories and limit leave out of the response); an empty result is not charged.

get_credit_balance

ChatGPT
Return the credits left on your account. Call it before starting a run that can spend many credits — a large generate_lead_list, or a loop of finder calls over hundreds of rows — so you can compare the estimated cost with what is left and ask the user before going ahead. Returns credits (finder credits left, used by every lookup tool and by lead lists), verifier_credits (credits left for verify_email — absent when the account is on unified pricing, where a single credits pool pays for everything) and has_capacity. Read them by plan: when has_capacity is null, the account buys credits and credits is exactly what it can spend. When has_capacity is true or false, the account is on a usage-capacity plan: spending first draws on rolling usage windows that refill over time, and credits is only an extra balance on top, often 0 — so credits: 0 with has_capacity: true does not mean the run will be refused. has_capacity false means nothing is left right now; calls are refused until the windows refill. On a team, the balance and the capacity are the team owner's, shared by every member. Consumes no credits.

get_credit_usage

ChatGPT
Report how many finder and verifier credits were spent over a date range. By default it covers your own usage, broken down by day — or by month when the range is longer than 2 months — under items, with the totals under total. Set team to true for a per-member summary of the whole team under members; only the team owner may do that, anyone else gets a 403. Use it to explain where credits went or to compare spend with what is left (get_credit_balance). On a usage-capacity plan (has_capacity not null in get_credit_balance), spend absorbed by the usage windows may not show here. Consumes no credits.

get_lead_list_results

ChatGPT
Retrieve results from a lead list search started with generate_lead_list. Takes the hash returned by generate_lead_list. Polls the search status for up to 50 seconds. If the search completes within that time, returns the full data. If still processing after 50s, returns the current progress — call again to continue waiting. Consumes no credits: the run itself was charged by generate_lead_list.

list_contact_lists

ChatGPT
List the contact lists you can use: the ones you own plus the ones shared with your team. A contact list holds people you found and saved — not to be confused with exclusion lists, which hold company domains to keep out of searches. Returns each list's id, name, whether it is shared with your team and whether you own it. Use the ids with list_contacts, find_email and generate_lead_list (to save what they find into a list), create_monitor (lead_list_id) and the other contact list tools. Consumes no credits.

list_exclusion_domains

ChatGPT
List the domains held by an exclusion list, paginated. Each domain is returned with its own id — those ids are what remove_exclusion_domains needs, so call this first when you want to remove domains. Omitting list_id does NOT return only your global exclusions: it returns EVERY domain you can reach — your global ones, every list you own, AND every list shared with your team. To act on the global exclusions only, keep the entries whose excluded_domain_list_id is null. Always pass an explicit list_id before deleting anything. Results are ordered by domain name alone, so an entry can move between pages — re-read the page before using an id. Consumes no credits.

list_exclusion_lists

ChatGPT
List the exclusion lists you can use (lists you own plus lists shared with your team). An exclusion list holds company domains that should never show up in search results. Returns each list's id, name, owner and sharing status — use those ids with generate_lead_list (exclusion_list_ids / add_to_exclusion_list_id), with lookalike_search, or with the other exclusion tools. Consumes no credits.

list_monitors

ChatGPT
List the signal monitors you can see — yours and the ones teammates shared — most recent first. A monitor watches LinkedIn and job boards for one kind of buying signal and delivers each match as a signal. Returns each monitor's id, name, type, status (active or paused), filters, enrichment level, contact list, match cap and match_count. Use the ids with list_signals, update_monitor and delete_monitor. Consumes no credits.

remove_exclusion_domains

ChatGPT
Remove domains from an exclusion list, making them eligible for search results again. Takes domain ids, NOT domain names — call list_exclusion_domains first to resolve names into ids. Deletion is permanent: there is no undo and no soft delete. If none of the ids are yours to delete the call fails with a 403 rather than reporting a count of zero; if only some are, the others are dropped silently and the response does not say which. On a team-shared list you are allowed to delete domains added by teammates, so check the list before deleting. Ids obtained right after a large add_exclusion_domains may be stale. Consumes no credits.

reverse_email_lookup

ChatGPT
Identify the person behind an email address. Returns their LinkedIn profile URL; set with_profile to also return the full LinkedIn profile (name, job title, company etc.). Use it to put a name and a company on an address that arrived without context — an inbound reply, a CRM row, an imported list — and chain the returned LinkedIn URL into find_phone when you need to reach them by phone. It says nothing about deliverability: use verify_email for that. Costs 1 finder credit per contact found, or 2 finder credits when with_profile returns complete profile data; an address that matches nobody is not charged.

search_technologies

ChatGPT
Search the Findymail technology catalog by name and return up to 25 matches, each with its exact name, category and subcategory. Use it to check that a technology is known, and how it is spelled, before naming it in a generate_lead_list query ("companies using HubSpot") or filtering get_company_technologies on it. Rate-limited to 10 calls per minute. Consumes no credits.

verify_email

ChatGPT
Check whether an email address is deliverable, without sending anything to it. Use it on addresses that came from elsewhere (a CRM export, a spreadsheet, the user) before adding them to an outreach sequence. Addresses returned by find_email and generate_lead_list are already verified and do not need a second pass. Costs 1 verifier credit per address checked — verifier credits are a separate pool from the finder credits the lookup tools spend.

Capabilities

Writes

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/findymail"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/findymail)