AI agents for recruiters and staffing agencies that work the whole candidate list by email, text and phone on a sequence you wrote, ask your screening questions on the call or by text and keep the transcript, book the interview slot into your calendar, and run the same discipline at the employers you want as clients. One agent per desk, on your own mailbox and number.
7-day free trial · Cancel anytime
Recruitment outreach automation starts with a list: a CSV from your database, a segment from a connected CRM, or contacts pulled off a page in the Chrome side panel. The agent works it on a sequence you wrote: an email on day 0 from your own mailbox, a text on day 2 from your own number, a call on day 4, a second email on day 7 only to the people who opened and did not reply. Each step has its own delay and send condition.
Several mailboxes per agent rotate through the sends, each with warm-up and a health score out of 100, so a new desk's address does not burn the domain. A reply comes back over IMAP and is answered by the agent from your job spec and your answers, and the sequence stops on that candidate the moment they reply, book or opt out.
Write the screening questions into the call script: right to work, notice period, salary expectation, the shift pattern they can do, whether they hold the licence. The agent dials on your own Twilio number, speaks in real time using your provider's fastest model so it does not leave silences, asks the questions, answers the candidate's own questions from the job spec, and brings the recording and transcript back to the inbox against the candidate.
By text it does the same in fewer words. The text step carries the sending window, the texts-per-hour pace for the number and the Reply STOP line, and a reply of STOP is honoured on every channel at once. Answering and dialling are separate switches, so the same number can take the call-back.
Emails, texts, call transcripts and DMs land in one inbox, newest first, tagged with the channel and the agent, which for most agencies means the desk or the client. Filter it to one mailbox, one number or one agent when you are working a single role. A consultant invited into their own desk's agent sees that inbox and nothing else.
When a candidate says something the agent should not handle, a counter-offer, a complaint, a question about a contract clause, you reply in the thread yourself and the agent stands down on that conversation. Hand it back when you are done and the sequence resumes or ends, as you decide.
Each agent has its own funnel: how many candidates it reached, how many replied, how many visited the job page, how many booked a screen or an interview, and the value won against a goal with a fee attached. A channel table shows where each candidate first responded and the channel they were on when they booked, so you learn whether it is the text or the call that moves your candidates.
One row per candidate shows every message, reply, call and booking in the order it happened. When a client asks what you did on the search, the record is there.
Recruiting does not have a sourcing problem. It has a contacting problem. The search produces two hundred names, the consultant calls the first twenty, screens six, submits three, and the other hundred and eighty are never contacted at all, because a sixth touch on a candidate who did not answer the first five is the job that moves to tomorrow. The role fills or it does not on the strength of twenty calls, and the client business development that keeps the desk fed gets the hour that is left, which is usually none.
Bold Agent Kit gives each desk one agent that works the whole list by email, text and phone on a sequence you wrote, from your own mailbox and your own number. You give it the job spec as a document, the salary band and shift pattern as data points, your answers to "is it remote?" and "who is the client?" as Q&A pairs, the job page on your site it may link to, and instructions on how to talk and when to hand a candidate to you. It then reaches every name on the list, on the days and inside the hours you set, and stops on each candidate the moment they reply, book or opt out.
It screens what it reaches. Write the screening questions into the call script or the text step and the agent asks them: notice period, right to work, salary expectation, the licence, the shift. On the phone it speaks in real time on your Twilio number and keeps the recording and transcript against the candidate. By text the answers land in the inbox. The AI recruiter assistant does not decide who is good; it asks what you told it to ask and puts the answers in front of you.
It books the screen. Connect Google Calendar, Microsoft 365, Calendly or Cal.com and the agent offers real open slots and books the one the candidate picks, then confirms it in the reply. It cannot book a time that was not free. The same agent, or a second one, runs client business development at the employers you want, and a third re-engages the candidates in your database who have not been contacted since the last role closed.
Compliance is built into the campaign, not bolted on. Days, hours and time zone are set before anything sends. Every campaign email carries a one-click unsubscribe, a reply of STOP or an opt-out on one channel ends the sequence on every channel and cancels what is queued, and the agent replies once per thread with a daily cap per sender and hourly caps per number and mailbox. The plan is $25 a month or $200 a year for the whole agency, however many desks, with the carrier and the AI provider billing you at cost.
The candidates exist and the search found them. Four things happen between the longlist and the interview, and none of them is a sourcing problem.
Two hundred names, twenty calls, six screens. The other hundred and eighty are the placement that went to the agency across the road, because their consultant happened to call the right person on the right afternoon. The list was fine. It was never worked.
The same eight questions, asked forty times, for six candidates worth submitting. Notice period, right to work, salary, the shift. Every one is a ten-minute call and a note that may or may not make it into the system, and the consultant's afternoon goes on the thirty-four who were not right.
"Does Tuesday work?" "No, Thursday?" "Thursday morning is gone." Three days of back and forth for a thirty-minute screen, and the candidate who was keen on Monday has accepted somewhere else by the time the slot is agreed.
The desk is fed by employers who have roles, and reaching them is a sequence of touches nobody has time for while there is a role to fill. So the pipeline of clients dries up exactly when the current search ends, and the cycle starts again from nothing.
You build the desk's agent once with the job spec, the questions and the sequence. From then on it reaches, screens and books on its own, and only fetches you for the candidates you named.
An email on day 0 from your own mailbox, a text on day 2 from your own number, a call on day 4, a second email on day 7 to the people who opened and did not reply. Each step has its own delay and send condition, and nothing goes out on a day or at an hour you have not allowed. Bluesky, X, Reddit and WhatsApp DMs can carry the conversation for the candidates who live there.
On the call the agent speaks in real time on your Twilio number, asks the screening questions in the script, answers the candidate's from the job spec, and brings the transcript and recording back against the candidate. By text the answers land in the inbox. It asks what you told it to ask and does not improvise a question or an offer.
It is given the slots that are genuinely free in Google Calendar, Microsoft 365, Calendly or Cal.com, offers a few in the reply, and books the one the candidate picks. It cannot take a time it did not offer. The booking is confirmed in the thread and logged against the candidate, and a booking on any channel ends the outreach sequence on all of them.
Developers, designers and trades are often easier to reach as a DM than as an email to an address they stopped checking. Connect Bluesky, X, Reddit, WhatsApp or Telegram with your own app credentials and the same agent answers there, inside each platform's reply window, from the same job spec and the same questions. The handle is attached to the candidate's contact record, so a DM reply and a text reply from the same person are one thread, not two.
Every DM conversation lands in the same inbox as the emails and call transcripts, tagged with the platform. The agent reads the whole history before it replies, so a candidate who was sent the spec by email on Monday and asks about the salary on Bluesky on Wednesday is not asked to repeat themselves.
Each one answers a problem above, and each one runs while the consultant is on the calls that need a person.
Two hundred candidates get the email, the text, the call and the follow-up on the days you set, and the sequence runs to its end on every one of them unless they reply, book or opt out.
Notice period, right to work, salary, the shift, asked on the call or by text and stored as a transcript against the candidate. The consultant reads six answers instead of making forty calls.
Real open slots offered the moment the candidate says yes, one of them written to the calendar and confirmed in the thread. No email tennis, no three-day gap for a competitor to fill.
A business-development sequence at the employers you want, answered from your case studies and terms, with the discovery call booked into your calendar, running whether or not there is a role to fill this week.
Every send, reply, call, transcript and booking is recorded against a candidate and reported as a funnel, per agent, which means per desk or per client. This is the part that tells you where your candidates actually respond.
The funnel counts candidates rather than messages: how many the agent reached, how many replied, how many visited the job page, and how many booked a screen or an interview, with the median time from first contact to the booking. Goals carry a dollar value, so a placement marked won reports the fee against the campaign that produced the candidate, and a business-development run reports the clients it opened.
Below the funnel, a channel table shows where each candidate first responded and which channel they were on when they booked, so you learn whether it is the text on day 2 or the call on day 4 that moves your market. One row per candidate shows every touch, reply, transcript and booking in order, which is the record a client asks for when they ask what you did on the search, and the record a candidate's complaint is answered from.
Built from the cookie, email address and phone number attached to that contact, so activity months apart on different channels appears on one row.
Page goals and visit triggers require a one-line snippet on your website. Conversations, bookings and manually marking a deal won work without it.
Each of these is usually a separate tool or a separate person, and most are priced per seat, per contact or per minute. Here they are one plan for the whole agency.
| The job | What it normally takes | Bold Agent Kit |
|---|---|---|
| Email sequences from warmed mailboxes | A sequencer priced per seat, plus a warm-up tool | Included, with warm-up and health scores |
| Text campaigns and replies | An SMS platform on top of your carrier | Included, you pay only your carrier |
| Screening calls with a transcript | A consultant's afternoon, or a voice tool priced per minute | Included, on your own Twilio number |
| Booking the interview slot | Three days of email, or a booking link | Included, the agent books it |
| Following up the candidates who went quiet | A resourcer, or nobody | Included, until they reply or opt out |
| Business development at employers | The hour that is left, or a separate outbound tool | Included, the same agent or a second one |
| Opt-outs, hours and caps | A spreadsheet and good intentions | Included, enforced on every channel |
| The AI itself | Metered per message, per minute or per credit | Your own key, billed by your provider at cost |
Bold Agent Kit is $25 a month, or $200 a year, for the whole agency: every desk, every agent, every channel, no per-seat or per-contact charge, and no percentage of a fee. The only other costs are your carrier's, for texts and calls, and your AI provider's, typically a few dollars a month per agent.
The same agent covers all of these. Most agencies run one per desk or one per client, and a separate one for business development.
A role opens and the longlist is loaded from a CSV, a CRM segment or the side panel. The agent works it on a sequence: email, text, call, follow-up, inside the hours you set. It answers "what is the salary?" from the spec, asks your screening questions, books the screen into your calendar, and hands you the candidates who pass.
The employers you want as clients, worked with the same discipline. Build a list of hiring managers and owners from the side panel or a CSV, run a sequence at them from warmed mailboxes, and let the agent answer "what are your terms?" from your material and book the discovery call. Staffing agency lead follow-up that runs whether or not there is a role to fill this week.
The candidates in your database who were placed two years ago, or shortlisted and not chosen, or registered and never contacted again. Export the segment as a CSV and run a candidate re-engagement campaign: an email asking whether they are open to a move, a text for the ones who opened, a call for the ones who replied. The agent answers what comes back and books the catch-up call.
Connect the tools you already run and the agent works them like an assistant would: it takes appointments straight into your calendar, writes notes and outcomes back to the CRM, and reads your store to see what somebody has bought before it answers them.
Bookings are real, not guesses. It is only ever given slots that are genuinely free, and it can only take one it actually offered — so it cannot invent a time or double-book you. The same connection pulls a segment of past customers into a list when you want to run a campaign at them.
HubSpot
GoHighLevel
Housecall Pro
ServiceTitan
Jobber
Google Calendar
Microsoft 365
Calendly
HubSpot Meetings
Zoho Bookings
Acuity
Cal.com
Square
Stripe
Shopify
WooCommerce
Wix
SquarespaceEach agent uses a model on your own API key, in the same way it sends from your own mailbox and your own phone number. Sixteen providers are supported: OpenAI, Anthropic, Gemini, xAI and Mistral; DeepSeek, GLM, Kimi, Qwen and MiniMax; Groq, Together, Fireworks and OpenRouter; and Ollama or any other OpenAI-compatible endpoint you host yourself.
Bring your own credentials and you own the whole stack: your model provider, your mailboxes, your carrier, your CRM. We charge for the software and nothing else — no token markup, no metering, no credit packs — so your provider bills you direct at their rates, usually a few dollars a month per agent.
OpenAI
Anthropic
Gemini
xAI
Mistral
DeepSeek
GLM
Kimi
Qwen
MiniMax
Groq
Together
Fireworks
OpenRouter
Ollama
Any OpenAI APIA new address sending forty cold emails a day gets filtered, and takes the domain with it. Mailboxes that opt into warm-up send each other ordinary mail on a ramp, starting at five a day. The receiving mailbox finds each message, moves it out of spam if it landed there, marks it read and replies to a share of them.
Each mailbox carries a health score out of 100 — SPF, DKIM and DMARC on the sending domain, seven days of real sends and bounces, and where its own warm-up mail is landing — with the reasons listed. A mailbox whose bounce rate or spam placement passes a threshold is pulled out of campaign rotation automatically, before it damages the domain.
Three steps. A campaign is created as a draft and shows you every candidate and every step before it sends anything.
Name it after the desk or the client, choose a model and paste your API key. Then give it what a good resourcer knows: the job spec as a document, the salary band, location and shift pattern as data points, your answers to "is it remote?", "who is the client?" and "what is the interview process?" as Q&A pairs, and the job page on your site it may link to.
Write the instructions last: how it should talk, the screening questions to ask and in what order, what it must never promise, and the situations in which it should stop and fetch you. A counter-offer, a question about a contract clause and a candidate who passes the screen are the usual three.
Add the desk's mailbox with an app password or as an SMTP and IMAP host, and a second and third if the volume warrants it, so campaigns rotate between them. Add a number for texts and calls on your own carrier account, and the social platforms your candidates use, with your own app credentials. Each is a separate switch.
Then connect the calendar the screens are booked into, and the CRM if you have one, so the agent reads the candidate before it writes. Load the list: a CSV, a CRM segment, or contacts pulled off a page in the Chrome side panel.
Build the sequence across the channels you want, set the days, hours and time zone it may contact anyone, add the conditions on each step, and read the whole plan on the review screen: every candidate, every step, every date. Nothing leaves until you press Start.
Then watch the inbox. Replies are worked by the agent, screens are booked, the candidates who pass and the conversations you named come to you. Pause the campaign when the role fills and resume it, or start the next one from the same list.
One account runs unlimited agents, one per client, and nothing is shared between them. A knowledge base belongs to a single agent, and content from one agent is never used to generate replies for another.
Invite a client by email address and they get their own login, which opens onto the one agent you gave them: its conversations, its numbers, its analytics and a reply box. They cannot see your other clients, your API keys, your settings or your billing. Access is deny-by-default.
No seat tiers, no credit packs, no per-agent or per-channel pricing, and nothing held back behind an enterprise call.
| Agents | |
|---|---|
| How many | Unlimited — one per business, brand or job |
| Shared brain | The same identity, instructions and knowledge on every channel |
| Knowledge base | Documents, Q&A pairs, data points and pages of your own site |
| Isolation | One agent's content is never used to reply for another |
| Client seats | Invite a client into one agent, with their own login |
| Channels | |
| As many mailboxes per agent as you connect, rotated by campaigns | |
| Text messages | Nine providers — Twilio, Telnyx, SimpleTexting, Plivo, Vonage, Sinch, Bandwidth, MessageBird, ClickSend |
| Phone calls | Answering and dialling, with recordings and transcripts |
| Transcription | Google, or Deepgram nova-3 on your own key |
| Voices | Google, Amazon Polly and ElevenLabs, or your own ElevenLabs / Fish Audio account |
| Voice cloning | Record a voice in the browser and clone it on your ElevenLabs or Fish Audio key |
| Website chat | A widget on your site with your colour, image and greeting |
| Social DMs | Messenger, Instagram, WhatsApp, Telegram, LINE, Viber, Slack, Discord, Bluesky, Reddit, X |
| Inbox | |
| One list | Every channel's conversations together, newest first, tagged by channel |
| Filters | By channel, by mailbox or number, or by agent |
| Takeover | Reply yourself and the agent stands down on that thread |
| Hand back | Return the thread to the agent when you are done |
| Campaigns | |
| Multichannel sequences | Email, text and call steps in one campaign, each with its own timing |
| Send conditions | Send a step only if they opened, did not open, clicked or did not click |
| Schedule | The days, hours and time zone anything may be sent |
| Stop rules | A reply, a booking or an opt-out on any channel ends the sequence |
| Pacing | Per-mailbox and per-number hourly limits, pausable and resumable |
| Review before start | The full plan is shown; a campaign sends nothing until you start it |
| Templates | Email, text and call scripts, saved once and reused |
| Analytics | |
| Funnel | Reached, replied, visited, converted, plus value won and time to convert |
| By channel | Where each person started, and the channel they converted on |
| Pages | Which pages known people read, and how many of them converted |
| Journeys | One row per person: every message, visit, call and booking in order |
| Goals | A page rule or a manual "won", each with a dollar value |
| Triggers | A known person returns to a page — notify you, or send a follow-up |
| Open and click tracking | Recorded per campaign and added to that person's history |
| Integrations | |
| Calendars | Google, Microsoft 365, Calendly, Cal.com, Acuity, Zoho, Square, HubSpot Meetings |
| CRMs and commerce | HubSpot, GoHighLevel, Housecall Pro, ServiceTitan, Jobber, Stripe, Shopify, WooCommerce, Wix, Squarespace |
| Booking | Real open slots offered, and only an offered slot can be booked |
| CRM import | Import a segment of customers into a contact list |
| Deliverability and compliance | |
| Mailbox warm-up | Ramped sending across the network, with spam rescue and replies |
| Health score | SPF, DKIM, DMARC, bounce rate and inbox placement, with reasons |
| Auto-rest | A failing address is removed from rotation automatically |
| Unsubscribes | One click, honoured automatically, including on queued mail |
| Reply limits | One reply per thread, per-sender daily and per-channel hourly caps |
| Filtering | Bounces, out-of-office and no-reply addresses ignored |
| List building | |
| Contact extraction | Emails, phone numbers and social profiles, on any site. Unlimited. |
| Deep scan | Follows the pages of a domain most likely to carry contact details |
| Domain queue | Load a CSV of sites and work through them one at a time |
| Contacts and lists | Unlimited, by email or phone, with CSV import and export |
| AI | |
| Your own key | Sixteen providers, including any OpenAI-compatible endpoint you host |
| What we charge for AI | Nothing — no markup, no metering, no credit packs |
| On calls | Uses your provider's fastest model automatically |
| Account | |
| Platforms | Chrome side panel and web dashboard on one login |
| Members' forum | Six boards, visible only once you are signed in |
| Referral program | $50 a year per referral ($5 a month), recurring for the life of the account |
| Support | support@boldagents.co |
Billed monthly. No add-ons, no seat tiers. Cancel anytime.
$16.67 a month, billed once a year. No add-ons, no seat tiers. Cancel anytime.
They contact the whole longlist and bring you the candidates worth talking to. One agent per desk works a list by email from your own mailbox, text from your own number and phone calls on Twilio, on a sequence you wrote with its own days, hours and conditions. It answers the candidate's questions from the job spec, asks the screening questions you wrote on the call or by text and keeps the transcript, offers real open slots from your calendar and books the screen, and hands you the candidates who pass. The same agent, or another, runs business development at employers and re-engagement at your database. It stops on each person the moment they reply, book or opt out.
Yes, on your own Twilio number, as a step in the sequence or when a candidate calls back. Write the screening questions into the call script: notice period, right to work, salary expectation, shift pattern, licence. The agent speaks in real time, using your provider's fastest model so it does not leave silences, asks the questions, answers the candidate's from your material, and brings the recording and transcript back to the inbox against that candidate. It asks what you told it to ask and does not improvise. Answering and dialling are separate switches, and the voice can be Google, Amazon Polly or ElevenLabs, or one you cloned on your own ElevenLabs or Fish Audio key.
Email, from as many mailboxes as you connect, each with warm-up and a health score. Text messages, on your own number through Twilio, Telnyx or one of seven other carriers. Phone calls, dialling and answering, on Twilio. Website chat, for the candidate who lands on your job page. And eleven DM platforms with your own app credentials: Bluesky, X, Reddit, WhatsApp, Telegram, Messenger, Instagram, LINE, Viber, Slack and Discord. Every handle seen for a candidate, an email address, a phone number, a social account id, is attached to one record, so the agent reads the whole history before it replies, whichever channel the reply arrives on.
Three places. A CSV export from your database, imported into a contact list. A segment pulled straight out of a connected CRM such as HubSpot or GoHighLevel. Or the Chrome side panel, which extracts email addresses, phone numbers and social profile links from whatever page you have open, follows the pages of a domain most likely to carry contact details, and works through a CSV of sites one at a time. The side panel signs in with the same account as the dashboard, so contacts, lists, templates and agents are shared, and you can draft and send a single email from it without creating a campaign.
It gives you the controls compliance needs. A campaign has a sending window with days, hours and time zone, set before anything sends. Every campaign email carries a one-click unsubscribe, the text step carries the Reply STOP line, and an opt-out on any channel ends the sequence on every channel and cancels what is queued. The agent replies once per thread, respects a daily cap per sender and hourly caps per mailbox and number, and ignores bounces, out-of-office replies and no-reply addresses. Every message and transcript is stored with timestamps. Whether your sequence meets the rules in the jurisdictions you recruit in remains your responsibility, as it would with a resourcer doing the same work.
Yes, and it books rather than sending a link. Connect Google Calendar, Microsoft 365, Calendly or Cal.com and the agent is given the slots that are genuinely free for the week ahead. It offers a few in the reply, checks that the candidate's choice was among those it offered, writes the appointment to the calendar and confirms it in the same thread. It cannot book a time that was not free, invent one, or cancel an appointment it was not shown. A booking on any channel ends the outreach sequence on all of them, and the booking is logged against the candidate so the consultant sees it before the call.
Either, and agents are unlimited on one account. Most agencies run one per desk, with that desk's mailboxes, number, job specs and funnel, and a separate one for business development. Some run one per client, so a consultant or the client can be invited in with their own login onto that agent alone: its inbox, its numbers and its analytics, with no sight of the other agents, your API keys, your settings or your billing. A knowledge base belongs to one agent, and content from one is never used to reply for another, so two clients in the same sector can sit on one account.
The same as with one. The plan is $25 a month or $200 a year for the whole account, with a 7-day free trial, unlimited agents, unlimited contacts and lists, and no per-seat, per-agent, per-contact or per-message charge. The other costs are your carrier's, for texts and calls at their rates, and your AI provider's, on your own key with one of sixteen providers, billed at cost and usually a few dollars a month per agent. There is no percentage of a placement fee, and the Chrome side panel and the dashboard share the one login.
One subscription, unlimited agents. What changes is the knowledge base, the channels and the instructions.
Load the list, write the sequence and the screening questions, read the plan, press Start. Every candidate is reached, every answer is kept, and the screens land in your calendar.
7-day free trial · Cancel anytime