Ask an AI assistant (like ChatGPT or Claude) for the email address of a VP of Marketing and you'll get a confident response like firstname@[company].com. 76% of the time, it’s wrong.
That’s right, Hunter’s research on LLM email accuracy found 76% of the 1,604 email addresses that AI models with no live data connection gave us were wrong. Even GPT-5.5, the strongest model in the study, ended with a valid contact on only 19.5% of requests. Then we gave the same model Hunter's MCP, and that number jumped to 78%.
Hunter's MCP connects Claude, ChatGPT, Gemini, and other AI assistants to your Hunter account, so your AI uses verified contact data instead of guessing. It covers every stage of AI lead generation, from finding target accounts to launching your first sequence.
This guide covers 26 Hunter MCP lead generation use cases, grouped into six jobs, and three workflows to chain them together. Each one provides a prompt you can paste, what comes back, and how many credits it costs.
First things first - if you haven't set it up yet, connect Hunter's MCP.
What is Hunter's MCP?
Hunter's MCP is Hunter's official remote server for the Model Context Protocol.
It lets AI assistants like Claude, ChatGPT, and Gemini use your Hunter account from inside a chat. With it, your assistant builds lead lists, finds and verifies emails, enriches contacts, organizes leads, runs email sequences, and sends leads to your connected CRM.
MCP is an approach that Anthropic released in November 2024 to connect any AI assistant to the tools where data lives. Hunter's server sits in front of the Hunter API and exposes it as individual tools. You describe what you want in your own words, and the AI picks which tools to run and in what order (and more importantly, you don’t have to know what Hunter feature does what).
The official server at mcp.hunter.io exposes over 100 tools at the time of writing, covering email sequences, templates, saved searches, company lists, and CRM sync alongside finding and verifying emails.
It works with Claude, ChatGPT, Gemini, Perplexity, Claude Code, Cursor, Codex, and any other MCP-compatible client. It's included on every Hunter plan (Free included), and uses the credits you already have available.
How to connect
Setup is one server URL: https://mcp.hunter.io/mcp.
Clients that support it connect through a one-click OAuth flow - the same kind of login as “Sign in with Google” or “Sign in with Microsoft” on other software. OAuth lets your AI pass your Hunter API key to Hunter in an X-API-Key header, so all you need to do is be logged in to Hunter.
Set up guides
Full walkthroughs are available on the Hunter MCP setup page, plus guides for:
If you're on Claude's Free plan, note that it allows a single custom connector, so make Hunter the one you add.
The cost of each action
Your AI and Hunter will ask for your permission before it spends any credits.
Here’s how much each action costs in Hunter (from our Help Center) for platform plans:
Action | Cost |
|---|---|
Find companies (Discover), count emails, and save searches from the B2B database | Free |
Domain Search through the MCP | 1 credit per 1-10 emails returned |
Find emails (Email Finder) | 1 credit per email found |
Verify emails (Email Verifier) | 0.5 credits per verification |
Person, company, or combined enrichment | 0.2 credits per enriched email or domain |
Leads, lists, email sequences, CRM pushes, and account checks | Free |
A few key points about the Hunter MCP costs:
- You won’t pay if Hunter can't find or verify an address.
- Repeating the same search in the same billing period doesn't cost you twice, and
- Sending emails through Sequences uses no credits at all.
The Free plan comes with 50 credits a month, which is enough to run most of the prompts below a few times.
It also caps AI-powered B2B database searches at 10 a month and saves companies at 100, and leaves out premium filters like Technology. Data Platform plans (built for API use) work differently, with separate Search and Verification credit pools.
Task | What you can do | Cost |
|---|---|---|
Find companies from a plain-English brief, size the list, save searches, and organize company lists | Free | |
Pull contacts at a company, find one person's email, and pick a primary and backup contact per account | 1 credit per 1-10 emails (Domain Search), 1 credit per email found (Email Finder) | |
Check deliverability, enrich people and companies, and qualify inbound leads | 0.5 credits per verification, 0.2 credits per enrichment | |
Save contacts without duplicates, tag them, add custom fields, and bulk-move lists | Free | |
Build sequences and follow-ups, add recipients, and launch | Free (sending uses no credits) | |
Push leads to HubSpot, Pipedrive, Zoho, Zapier, or Salesforce | Free |
Build target account lists without spending a credit
Hunter's MCP turns a one-line description of your ideal customer into a list of real companies, and none of it costs credits.
Account research runs on Hunter’s B2B database, Discover, which is free to search.
Describe your ideal customer and get matching companies
Type the kind of company you sell to, the way you'd talk to a colleague.
The assistant turns it into Discover filters and returns up to 100 companies per page.
Describe your ideal customer and get matching companies
Hunter, find B2B SaaS companies in Germany with 51-200 employees that sell HR software.
I ran this exact prompt while writing this post, and it's a great lesson.
Hunter came back with 2,067 matches. The filters it applied were Germany, 51-200 employees, and a keyword list matched on "any": B2B, SaaS, HR software, and three variants.
That "any" is the catch. The top result was a printing machinery maker, because "B2B" matched.
Real fits like Flip (an AI employee app) and ValueNet (an employee benefits platform) were further down the list.
The solution took one follow-up: drop the generic words and keep the specific ones.
Refine the search with specific keywords
Run that again, but only use keywords that describe HR software, including HR software, HR tech, and employee experience. Show me the filters you applied before the results.
The response also includes a permalink that opens the same search in Discover, so you can fine-tune filters by hand if the chat version gets close but not quite there.
Tools used: Discover · Cost: Free
Find companies that use a competitor's tool
The MCP can help your AI find companies based on the technologies it uses. It's a premium filter in Hunter’s B2B database, so you'll need a paid All-in-One plan.
Every company in my German search came back with its tech stack attached, so a switcher list is a one-line ask.
Find companies that use a competitor's tool
Find marketing agencies in the UK with 11-50 employees that use HubSpot. Show me the technology filter you applied before the results.
Swap HubSpot for a competitor's product, and the list becomes accounts that already pay to solve the problem you solve.
Alternatively, if you integrate well with another product, or the existence of a technology indicates a company’s maturity or buying need, this is a great opportunity to leverage.
For a tighter shortlist, ask your AI to cross-check which of them are hiring for the role you sell to.
Tools used: Discover · Cost: Free
Turn a rough brief into a prospecting plan first
Big, fuzzy asks go better when your AI plans before it acts.
Hunter has a planning tool for this, and it's free to use.
Turn a rough brief into a prospecting plan first
Hunter, I want to reach heads of growth at fintech startups in the UK. Plan the steps and estimate the credit cost before you run anything.
What comes back is a step-by-step plan (companies, people, then verification) and the first action to take.
By default it gathers contacts for review in a table.
It only saves them to your Hunter Leads if you explicitly ask.
Tools used: Discover, Domain Search, Email Verifier (in the order the plan sets) · Cost: Free (each step it runs charges its own credits)
Size the list before you spend credits
Counting contacts costs nothing, so do it before you reveal any email address for a lead.
Hunter tells you how many personal and generic addresses it has for each company, plus a total across every match.
Size the list before you spend credits
Check how many personal email addresses Hunter has at each of these 20 companies. Don't reveal any yet. Rank them by coverage, split them into companies where Domain Search will do the work and companies I'd have to research by hand, and warn me if more than a third land in the second group.
You get a table of companies with personal, generic, and total counts.
Personal addresses are the ones tied to a named person, which are the ones you can prospect.
The warning at the end will help you to understand the degree of manual research you’ll need to do. If a third of a segment needs manual research, weigh up if that’s the right plan before you commit to prospecting it.
Tools used: Discover · Cost: Free
Save a search and rerun it later
Once a search is dialed in, save it so you don't rebuild it from memory next month. Just ask your AI.
If you do want to rerun the search inside the Hunter web platform, your saved searches show up in the Discover B2B database.
Save a search and rerun it later
Save this search as "UK fintech 50-200" so I can rerun it next month.
Tools used: Discover (saved searches) · Cost: Free
Organize accounts into company lists
Company lists keep target accounts in one place before you look for people.
Your AI can work with Hunter to create lists and folders, add or remove companies, favorite lists, and bulk copy or move companies between them.
Organize accounts into company lists
Create a company list called "Q4 targets" inside a folder named "EMEA" and add the top 15 companies from this search.
You get the new list back with its company count, ready for the next step.
Tools used: Company lists · Cost: Free
For similar prompts, visit list-building prompts in the official Hunter AI prompt library.
Find decision makers and their email address
Hunter's MCP works with your AI to find decision makers at a company, with their email address, title and confidence score, instead of inventing emails that fit a pattern.
These actions cost credits, so it pays to filter tightly. Here’s how you can do this.
Pull every contact at a company
Domain Search returns the people Hunter knows at a company.
You can filter by department, seniority, and email type, or only keep results that include a job title.
Pull every contact at a company
List the senior and executive people in sales at acme.com. Include names, titles, and confidence scores, and skip anyone without a job title.
You get a table of names, positions, emails, and confidence scores.
Nothing is saved unless you ask for it.
Tools used: Domain Search
Cost: 1 credit per 1-10 emails returned, charged only when emails are found. A result with 25 emails costs 3 credits.
Show only addresses Hunter has seen published
Regular Domain Search can include inferred addresses, meaning emails Hunter derived from a company's email pattern.
The "found" version returns only addresses Hunter has seen published on the public web, with nothing inferred.
Use it when you need to be extra careful, like before a high-stakes email to a single executive.
For general list building, the regular search plus verification gives you more coverage.
Show only addresses Hunter has seen published
Use only emails Hunter has seen published on the web for marketing people at acme.com. No pattern-based guesses.
Tools used: Domain Search (found emails only) · Cost: 1 credit per 1-10 emails returned
Find one person's email
When you know who you want to contact, give your AI a name and a company.
Find one person's email
Find the email for Jordan Lee, VP Marketing at Acme. Tell me the confidence score and whether it's verified.
You get one email address, a confidence score, and the verification status.
If Hunter can't find it, the assistant says so instead of making one up.
Tools used: Email Finder · Cost: 1 credit per email found
Decide how many people to email at each company
The right number depends on how big your total addressable market is.
In our study of 25,138 Hunter Sequences campaigns, companies where two or more people were contacted replied 3.35% of the time, against 1.61% when only one person was.
Each extra email works less well, though: the reply rate per person drops from 1.61% for a solo contact to 1.36% per person in a trio.
That means you can approach prospecting a buying committee in two ways:
- Small market, scarce accounts
- Contact 2-3 people per company.
- Pick non-executives from different departments, who replied at 2.30% against 1.65% for two people in the same department.
- Add the company's info@ inbox as a good alternative.
- Large market, more prospects than sending capacity
- Email one best-fit person per company and send to a second contact after your first contact’s sequence ends.
- That squeezes the most total replies out of your volume, and the re-engage workflow further down uses the backup when the first person goes quiet.
The State of Email Outreach 2026 points the same way from the recipient's side: contacting one or two people per company replied at 5.1%, against 3.5% for three or more.
Decide how many people to email at each company
These 25 accounts are my whole market. For each one, build a two-person plan: the best-fit non-executive for outbound prospecting, plus either the company's info@ inbox or a second named person from a different department. Verify the named contacts and give me a table: company, recipient 1 (name, title, email, why them), recipient 2, and confidence.
You get a two-person plan per account, ready to save and add to a sequence.
Good to know: Generic inboxes unsubscribed almost twice as often as named people in the study (2.06% vs 1.18%), because whoever triages info@ deals with your email one way or another.
Tools used: Domain Search, Email Verifier
Cost: 1 credit per 1-10 emails returned per company, plus 0.5 per verification
More prompts like this: prospecting prompts in the Prompt library.
Verify and enrich before sending an email
Hunter's MCP can check whether an address is safe to use and fill in details about the person and company, all in the same chat in your AI. Verification costs 0.5 credits. Enrichment costs 0.2.
Check an email address before you send
Paste any address and ask if it's safe.
Your AI will verify the email address and explain the result in plain English.
Check an email address before you send
Is jordan.lee@acme.com safe to send to? If it's accept-all, tell me the score and what you'd do with it.
You get:
- A status (valid, invalid, accept-all, disposable, webmail, or unknown) and
- A confidence score.
Good to know:
- Put accept-all emails into separate lists. In our tests, accept-all addresses bounced about 27x more often than valid ones.
- Don't bother verifying generic emails like info@ or sales@. The result usually comes back accept-all, so while you can email this address, don’t spend credits verifying.
Tools used: Email Verifier
Cost: 0.5 credits, charged only for valid, invalid, or accept-all results
Enrich a contact or a company
Give your AI an email address and it gives you the details behind it:
- Name
- Title
- Employer
- Location, and
- Social profiles.
Give it a company name or website, and it returns its industry, size, location, technologies, and funding.
Enrich a contact or a company
Enrich priya@northwind.io. Tell me her role, the company's size and industry, and what tech they use.
Good to know:
- If you want both profiles at once, combined enrichment takes an email address or a LinkedIn handle.
- Company enrichment returns no personal data, so it's the safe default when you only need account context.
Tools used: Lead Enrichment, Company Enrichment, Combined Enrichment
Cost: 0.2 credits per enriched email or domain, charged only when the core fields come back (for a person: email, full name, and position)
Qualify an inbound lead in one message
A demo request comes in, and you want to know in 30 seconds whether it's worth a fast reply. One prompt will conduct three checks for you.
Qualify an inbound lead in one message
A demo request just came in from sam@brightpath.co. Verify the address, tell me about the company, and list two or three senior people I should loop in.
Good to know: Behind that prompt Hunter will verify, enrich the company, and find senior people in the company.
All you see is a short brief telling you if the email address is valid or not, company size and industry, and who else you should consider connecting with in the buying committee.
Tools used: Email Verifier, Company Enrichment, Domain Search · Cost: about 1.7 credits (0.5 + 0.2 + 1)
Triage a week of signups
Paste every trial or demo signup from the week and get your AI to sort them before anyone on your team replies.
Triage a week of signups
Here are this week's 40 trial signups. Enrich each one, score them against my ICP (B2B SaaS, 20-200 employees, selling to sales teams), and sort them into three groups: reply personally today, leave on the standard onboarding flow, or no action. Flag any address Hunter returned nothing for.
Good to know: The flagged email addresses should be investigated - a signup with no match is often a throwaway account.
Tools used: Combined Enrichment · Cost: 0.2 credits per enriched email, charged only when the core fields come back, so 40 signups cost 8 credits at most
Book meetings before an event
It's tempting to start event outreach after the event, when everyone's inbox is full. The exhibitor list is usually public weeks earlier, which is when meetings get booked.
Book meetings before an event
I'm attending a SaaS conference in Berlin next month. Here's the exhibitor list. For each company, find one verified contact in partnerships or sales leadership and draft a two-line note that mentions the event and asks for 15 minutes there.
You get one contact per exhibitor and a short draft for each, ready for a sequence that finishes before you travel.
Tools used: Domain Search, Email Verifier · Cost: about 1 credit per company, plus 0.5 per verification
Clean up an event list
Conference lists often have names and companies, no emails. Upload the CSV to your AI and let it work through the rows.
Clean up an event list
This CSV has a first name, last name, and company from a trade show. Find each person's work email, add the email and confidence score as new columns, and flag any rows you couldn't match.
You get the original list back with emails, scores, and a clear list of misses.
Good to know: You can ask your AI to verify the list of email addresses if you receive these through a lead magnet or you have access to a conference portal that provides email addresses of event attendees.
Tools used: Discover (domain lookup), Email Finder · Cost: 1 credit per email found. Email Finder checks deliverability before it returns an address, so you don't need to pay for a separate verification
For more prompts like this, visit our inbound lead prompts in the official Hunter AI prompt library.
Keep your leads organized as you go
Everything you find in your AI can go into Hunter Leads without a single CSV export, and none of the lead management tools cost credits.
The one thing to watch is the difference between saving and overwriting.
Save contacts without creating duplicates
Ask your AI to skip anyone already in your account when you perform searches. It checks each email first and only creates the leads that are new.
Save contacts without creating duplicates
Save these 12 contacts to my "Q4 targets" leads list. Skip anyone who's already in Hunter and don't change their existing details.
You get a short report: how many were created and which ones already existed and were left alone.
Good to know: Be precise with your wording - "save" and "skip existing" map to a tool that never overwrites. "Update" maps to one that does, and an overwritten field can't be restored from the chat. If you have notes on a lead that you care about, say "don't overwrite" explicitly.
Tools used: Leads · Cost: Free
Keep lists tidy
Lists grow fast once AI can create them in a sentence. Good news: you can also clean lists in your AI. Rename lists, merge duplicates, favorite the ones you use, and group the rest into folders.
Keep lists tidy
Merge "Q3 fintech" into "Fintech 2026", then put every list with "event" in the name into a folder called "Events".
Tools used: Leads (lists and folders) · Cost: Free
Tag leads and add custom fields
Tags and custom attributes let you slice leads later without building a new list for every angle.
A tag like "conference-2026" or a custom field like "partner tier" takes one line to set up.
Tag leads and add custom fields
Create a custom field called "Partner tier" and tag everyone from yesterday's import with "conference-2026".
Tools used: Leads (custom attributes and tags) · Cost: Free
Filter and bulk-edit
A real time-saver that AI helps with is a task involving a lot of data. These bulk-type moves are where you describe a group of leads, and your AI finds and moves them in one.
Filter and bulk-edit
Find every lead in "Q3 outreach" tagged "no-reply" and move them to a list called "Re-engage". Show me the count before you move anything.
Good to know: Asking for the count first is a good habit for any bulk action, and essential for bulk deletes. AI will happily delete 400 leads if that's what your sentence says.
Tools used: Leads (bulk move) · Cost: Free
More prompts like this: list planning prompts in the Prompt library.
Build and launch outreach sequences from the chat
Your assistant can build a full Hunter Sequences email campaign from the first draft to launch, without you opening the web platform. Sending through Sequences uses no credits.
Starting a sequence sends real email to real people. Hunter makes the assistant ask for your explicit confirmation before any sequence starts. Read the steps, check the recipient count, and only then say yes.
Build a sequence and its follow-ups
Every new email sequence starts as a draft that sends nothing.
AI writes the first email, adds follow-ups with wait days between them, and sets the sending schedule.
Build a sequence and its follow-ups
Create a draft sequence called "Q4 targets - EMEA". Three steps: an intro email, a follow-up on day 3, and a short bump on day 7. Send Monday to Friday, 9am to 4pm, with open tracking off. Don't start it.
Good to know:
- You get the draft with each step listed so you can review the copy in the chat.
- Manually review and edit the drafts so that they read true to how you write
- Any merge fields (meaning where you pull in a contact’s name, company, job title, technology, custom attribute etc.) need a fallback, written as {{first_name:"there"}}. If you miss this, they will render empty and create issues like an email that starts with "Hi ," that reads like a mail merge gone wrong.
- Note that changing the intro's subject line copies the new subject down to later steps that still share it, so reread every step after a subject edit.
Tools used: Sequences · Cost: Free
Give AI your copy rules up front
AI will write whatever you ask for, and will always go for the most average email possible.
So, give AI rules to improve its writing.
These come from the same State of Email Outreach data, based on 31 million emails sent through Hunter in 2025:
- Send three emails: three messages lifted total replies by 106% (6.8% vs 3.3%). Going past three lowered the average again.
- Leave open tracking off: sequences without tracking replied at 7.4%, against 4.4% with it on.
- Personalize with two custom attributes: subject lines with two opened at 40.2% vs 35.4% with one, and bodies with two replied at 5.6% vs 3.6% with none.
- Ask one low-pressure question: decision makers in the survey preferred open-ended asks like "Open to learning more?" over a meeting request in the first email.
- Edit drafts by hand: sequences where the sender edited some emails before sending beat fully automated ones by 18% (5.2% vs 4.4%). Let the assistant write the first draft, then edit the emails that matter most.
Good to know:
- Use Hunter’s free Outreach Planner for guidance on what emails to send year round
- Tell your AI which details are merge fields and which have to come from you.
- Merge fields like {{company:"your team"}} fill from your Hunter leads.
- The number that proves your offer, or the customer you name, can only come from you, and an AI left to fill that gap will invent one.
Give AI your copy rules up front
Write the three emails for my "Q4 targets - EMEA" draft. Use two merge fields in each subject and body, end each email with one low-pressure question, and keep open tracking off. Wherever you need a result or a customer name, leave a [[placeholder]] for me instead of making one up.
Tools used: Sequences · Cost: Free
Pick a follow-up pattern instead of a generic bump
"Just checking in" is the default follow-up, and it gives the reader nothing new.
Different situations call for different follow-ups, and AI can create those when you ask for it:
- A breakup email acknowledges the silence and asks permission to close the file
- A wrong-person note says in one line why you wrote and asks who owns the topic
- A proposal chase asks for a rough decision timeline instead of repeating the pitch, and
- A meeting recap lists decisions, owners, deadlines, and the one item still open.
Pick a follow-up pattern instead of a generic bump
Rewrite step 3 of "Q4 targets - EMEA" as a breakup email. Acknowledge they're probably busy, offer one low-risk option, and ask whether I should close the file. Keep it under 80 words.
The follow-up section of the Prompt library has full prompts for each of these patterns, plus webinar and post-event follow-ups.
Tools used: Sequences · Cost: Free
Add or remove recipients
Recipients can come from your Hunter leads or straight from a list of email addresses.
Removals work the same, which helps when someone replies on another channel and shouldn't get the next step.
Add or remove recipients
Add everyone in the "Q4 targets" leads list to the "Q4 targets - EMEA" sequence. Then remove sam@brightpath.co, we're already talking.
Tools used: Sequences · Cost: Free
Turn opens and clicks into a warm list
People who opened or clicked an email but never replied know your name. Another follow-up rarely moves them, but a different reason to talk can.
Turn opens and clicks into a warm list
From my "Q3 outreach" sequence, pull the recipients who opened or clicked but never replied. Re-verify their addresses, then draft a short invite to our customer roundtable, personalized to each person's role, with one question as the only call to action. Set it up as a new draft sequence and don't start it.
Good to know: This works on sequences that have tracking on, which is the trade-off from the copy rules above. Tracking costs replies, but it's the only way to know who engaged. Link click tracking also needs a paid plan.
Tools used: Sequences, Email Verifier · Cost: 0.5 credits per re-verification
More prompts like this: campaign prompts in the Prompt library.
Sync leads with your CRM and other tools
Leads you build in the chat can go straight to your CRM, as long as it’s already connected through Hunter integrations. These tools are free.
Push leads to your CRM
HubSpot, Pipedrive, Salesforce, Zoho, and Zapier connections support lead syncing. Ask your assistant to show the field mapping first, so “position” in Hunter doesn’t land in the wrong CRM field. If something’s off, fix the mapping in Hunter’s integration settings.
Push leads to your CRM
Show me how Hunter fields map to my HubSpot fields. If it looks right, push the "Q4 targets" leads list to HubSpot.
Good to know: The push asks for confirmation before it writes anything, then runs in the background, so check your CRM shortly after. You can push a whole list or up to 100 specific leads per request.
Tools used: Integrations (connected CRM) · Cost: Free
3 workflows to chain it all together
Chained prompts are where the MCP really shines, because your AI carries context from one step to the next: the companies it found become the domains it searches, which become the leads it saves. The dashboard can't do that without you copying things between screens.
Credit totals below are examples.
Your numbers depend on how many emails each company returns.
From ideal customer to a ready-to-launch email sequence in one chat
You have an ICP and an empty block of time.
The goal is a reviewed draft sequence with verified contacts by the end of it.
From ideal customer to a ready-to-launch email sequence in one chat
Hunter, find 10 UK fintech companies with 51-200 employees. For each, find up to 3 senior people in marketing or growth. Verify any email addresses, save everyone new to a leads list called "UK fintech Q4", then build a 3-step draft sequence for that list. Show me the credit estimate before you search for people, and don't start the sequence.
What the AI does:
- Finds matching companies in Hunter’s B2B database for free and shows you the list.
- Counts available contacts per company (free) and estimates the credit cost.
- Waits for your approval, then runs a filtered search per company.
- Verifies the email addresses.
- Saves new contacts to the leads list, skipping existing ones.
- Drafts the sequence and stops for your review.
Tools used: Discover, Domain Search, Email Verifier, Leads, Sequences · Cost: 10-20 credits for the searches, plus 0.5 per verified accept-all address
The Monday list refresh
Saved searches go stale the moment new companies match them.
This chain reruns one and only spends credits on what's new.
The Monday list refresh
Rerun my saved search "UK fintech 50-200". Show me the query you'll use first. Compare the results to my "UK fintech Q4" company list, add only the new companies, and find senior marketing contacts at those new ones only.
What your AI does:
- Reads the saved search and shows you the rewritten query to confirm.
- Runs it and compares the results to your company list.
- Adds the new companies to the list.
- Runs a search on the new companies only.
- Saves new leads without touching existing ones.
Tools used: Discover (saved searches), Company lists, Domain Search, Leads · Cost: 1-2 credits per new company. Weeks with no new matches cost nothing.
Re-engage cold account
My team runs a version of this for link building, and it works the same way for lead generation.
Someone we emailed months ago never replied, but the company is still a good fit.
Often the original contact changed roles, or was the wrong person to begin with.
What works is finding a different contact at the same company.
We built this as a standalone automation.
With the MCP it's one prompt to our AI:
Re-engage cold account
Find leads I emailed more than three months ago who never replied. For each company, find one different contact in sales, marketing, or growth, ranked by how close their role is to buying what we sell. Save them to a new list called "Re-engage Q4" and skip anyone already in my leads.
What the AI does:
- Lists leads with no reply after three months and groups them by company.
- Runs a filtered search at each company.
- Picks the best-fit new contact per company by role.
- Saves them to the new list, skipping duplicates.
Tools used: Leads, Domain Search · Cost: about 1 credit per company
It works the same for closed-lost deals, with one extra step. Contacts from deals you lost a year ago may have left, so verify first and replace only the ones that fail:
Re-engage closed-lost deals
Here are 30 contacts from deals we lost last year. Verify each one. Keep the valid ones, and for every invalid address, find whoever now holds that role at the company. Return the two groups separately.
Keep the groups apart. The email to someone new in the seat needs a different opening than the one to the person who said no last year.
Tools used: Email Verifier, Domain Search · Cost: 0.5 credits per verification, plus about 1 credit per replacement search
What Hunter's MCP doesn't do yet
A few things still live in the Hunter web platform at time of writing:
- Signals: Hunter's company monitoring alerts (funding rounds, job openings, and company changes) aren't exposed as MCP tools. TechLookup isn't on this list because it opens a Discover search with the Technology filter, which the MCP can already run on paid plans.
- Bulk tasks: for lists of thousands, the app's bulk finder and verifier are still the faster route. The MCP works through rows one by one.
- Connecting a new inbox: the first connection of a sending account happens in the dashboard. After that, the assistant can use it.
- Exact saved-search reruns: reruns are rebuilt from the stored filters as a fresh query, so review them before running.
Tips for better results
Every prompt in this guide has the same four parts. Once you can spot them, you can write your own instead of copying ours.
- The task: one specific piece of work, stated first. "Find senior marketing people at 10 UK fintechs" is a job. "Help me with outreach" is a topic, and AI will fill the gap with guesses.
- The inputs: the list name, domain, ICP, or numbers only you know. Anything you leave out, the AI decides for you.
- The output shape: a table with named columns, a ranked list, or a draft held for review. Naming the shape stops it from returning a wall of prose.
- The stop condition: where it hands control back to you. "Show me the credit estimate before you search" and "don't start the sequence" are both stop conditions, which is why nearly every prompt here ends with one.
Take the first workflow apart and all four are there:
- The task is a ready-to-launch sequence for UK fintech companies.
- The inputs are the company count, the size range, and the departments.
- The output is a saved leads list plus a draft sequence, and
- The stop conditions are the credit check and "don't start the sequence."
A few tweaks help on top of that:
- Start prompts with "Hunter, help me..." so your AI reaches for Hunter instead of its training data.
- Name the department and seniority.
- Ask for confidence scores every time, and
- Count contacts before you reveal them.
When you want a head start, the Prompt library has ready-made prompts built the same way, sorted by role and task.
FAQ
How do I use Hunter’s MCP for lead generation?
Describe your ideal customer in plain English, and the assistant runs the whole chain in one chat. It starts with matching companies from Discover and pulls the decision makers at each one with Domain Search. Then it verifies their emails, saves them as leads, and drafts a sequence. Ask it to show a credit estimate before it searches for people, and to stop before starting the sequence.
Can Hunter’s MCP build a lead list from a description of my ideal customer?
Yes. Describe the companies you sell to, and the assistant turns that into Discover filters and returns up to 100 companies per page. Discover searches are free. Ask it to show the filters it applied, since generic keywords like “B2B” can pull in companies that don’t fit.
How many credits does a lead generation workflow use?
Finding companies and counting contacts is free. A workflow that finds 10 companies and up to 3 contacts at each usually costs 10-20 credits for the searches, plus 0.5 credits per address you verify. You’re never charged when Hunter finds nothing.
How do I find decision makers at a target company?
Ask for people by department and seniority, for example “senior people in marketing at acme.com.” The assistant runs a filtered Domain Search and returns names, titles, emails, and confidence scores. If you already know the person, Email Finder takes their name and company and returns one address.
Should I contact more than one person at each company?
It depends on the size of your market. In our study of 25,138 Hunter Sequences campaigns, companies where two or more people were contacted replied 3.35% of the time, against 1.61% with one contact. For a small market, contact two or three people from different departments. For a large one, email one best-fit person and keep a backup.
How accurate are the leads Hunter’s MCP finds?
In our test, 45% of the addresses GPT-5.5 gave on its own were invalid. With Hunter connected, that dropped to 3%. Email Verifier then gives each address a status and a confidence score. For high-stakes emails, ask for addresses Hunter has seen published, with nothing inferred from email patterns.
Can I qualify inbound leads with Hunter’s MCP?
A single prompt verifies the address, enriches the company, and finds two or three senior people to loop in, for about 1.7 credits. For a batch of signups, the assistant can enrich each one and sort them by fit with your ICP for 0.2 credits per enriched email.
Can Hunter’s MCP turn my leads into an outreach sequence?
Yes. The assistant builds a draft sequence with follow-ups and adds your saved leads as recipients. Nothing sends until you explicitly confirm the start, and sending through Hunter Sequences uses no credits.
Can I push the leads it finds to my CRM?
If your CRM is connected to Hunter, yes. HubSpot, Pipedrive, Zoho, Zapier, and Salesforce connections support lead syncing. The assistant can show you how Hunter’s fields map to your CRM first, and it asks for confirmation before pushing up to 100 leads per request.
Can I run lead generation with Hunter’s MCP on the Free plan?
Yes, with limits. The Free plan includes 50 credits a month, 10 AI-powered Discover searches, 100 saved companies, one connected email account, and up to 500 recipients per sequence. Premium Discover filters like Technology need a paid plan.
Try your first workflow today
Pick the first workflow above, swap in your own ICP, and run it with the "don't start the sequence" line left in. It costs a handful of credits and shows you the whole loop in one sitting.
Connect Hunter's MCP to get started. If you're new to Hunter, create a free account first. It takes a minute and comes with 50 credits a month.
Send cold emails with Hunter