reply-io-community #announcements

Channels

deactivateduser

05/06/2026, 10:46 AM

👋 Welcome to the Reply.io API community! This Slack is the place for anyone building with or connecting to Reply — whether you're enjoying the API directly, wiring up flows in n8n or plugging Reply into your AI agent through the Reply MCP server. Whether you write code, drag-and-drop in n8n, or just want to talk to other people automating their outreach — you're in the right place. Quick tour of the channels: 📢 #C0B02EDRC1X — Release notes, new endpoints, breaking changes, n8n / MCP updates. Check here first before asking "is X live yet?" 💬 #C0B02SVE8QH — Feature requests, ideas, "it would be great if Reply could…", rough edges in the API or n8n node. The dev team reads this. Upvote with 👍 on the messages you care about. 🆘 #C0B22AGCP8U — Stuck on something? Unexpected error, n8n workflow not firing, MCP tool returning weird results? Drop a thread here. To get help quicker, please include: • What you're trying to do (the goal, not just the symptom) • The endpoint / n8n action / MCP tool involved • The request you sent and the response/error you got (redact secrets!) A few house rules: 🔒 Never paste your API key, OAuth token, or other secrets in public channels. If you need to share a real example, DM the responder. 🧵 Keep replies in threads so the channels stay easy to scan. 🌍 Be kind, assume good faith, help each other out.

🙌 1

Dasha

05/14/2026, 4:53 PM

📢 New Webhook Event:
linkedin_account_alerts
Hey everyone! 👋 We've shipped a new webhook event that lets you stay on top of LinkedIn account health in real time. Instead of finding out after a sequence stalls, you'll get notified the moment something needs attention. What you'll be alerted about: • 🔌 Sales Navigator disconnected - reconnect needed to keep InMail steps running • 💬 InMail credits depleted - the account can't send InMails until credits renew • 🔒 Weekly connection request limit reached - sending resumes after LinkedIn resets the quota Why this matters: Build automations that auto-pause sequences, ping the account owner in Slack, open a ticket, or trigger any custom recovery flow, no more silent failures. Example payloads: Sales Navigator disconnected

Copy code

{
  "event": {
    "date": "2026-05-13T10:45:46.3926107Z",
    "id": "22ff637a-fd81-40ba-b597-0c8f26189dca",
    "type": "linkedin_account_alerts",
    "user_id": 12345,
    "team_id": 67890,
    "subscription_id": 492
  },
  "workspace_id": 67890,
  "workspace_name": "Default Team",
  "linkedin_owner_name": "Jane D.",
  "linkedin_owner_email": "<mailto:jane.d@example.com|jane.d@example.com>",
  "linkedin_id": 262,
  "linkedin_name": "Jane Doe",
  "linkedin_account_alert": "Sales Navigator is disconnected. Please reconnect your LinkedIn account to continue sending InMail steps.",
  "sequence_fields": {}
}

InMail credits depleted

Copy code

{
  "event": {
    "date": "2026-05-13T10:47:16.6543506Z",
    "id": "7227258a-8624-410a-bf6c-bd5197810d66",
    "type": "linkedin_account_alerts",
    "user_id": 12345,
    "team_id": 67890,
    "subscription_id": 492
  },
  "workspace_id": 67890,
  "workspace_name": "Default Team",
  "linkedin_owner_name": "Jane D.",
  "linkedin_owner_email": "<mailto:jane.d@example.com|jane.d@example.com>",
  "linkedin_id": 262,
  "linkedin_name": "Jane Doe",
  "linkedin_account_alert": "InMail credits are used up — this account can't send InMails until credits renew.",
  "sequence_fields": {}
}

Weekly connection limit reached

Copy code

{
  "event": {
    "date": "2026-05-13T10:47:27.4486303Z",
    "id": "654258f1-a426-4d3c-96a2-b3da5f78af29",
    "type": "linkedin_account_alerts",
    "user_id": 12345,
    "team_id": 67890,
    "subscription_id": 492
  },
  "workspace_id": 67890,
  "workspace_name": "Default Team",
  "linkedin_owner_name": "Jane D.",
  "linkedin_owner_email": "<mailto:jane.d@example.com|jane.d@example.com>",
  "linkedin_id": 262,
  "linkedin_name": "Jane Doe",
  "linkedin_account_alert": "Weekly connection request limit reached. Sending new connection requests will resume after LinkedIn resets the quota.",
  "sequence_fields": {}
}

📚 Full docs: https://docs.reply.io/webhook-event-payloads#linkedin_account_alerts Would love to hear what you build with it, drop a comment or share your workflow! 🚀

🎉 3

🔥 1

Dasha

05/18/2026, 12:37 PM

🛠️ New: Webhooks Management Page (v3) Hey folks! 👋 Following up with another shipment, we've added a dedicated Webhooks page where you can manage your v3 webhook subscriptions directly from the UI. No more juggling subscriptions through API calls only. Availability: Currently rolled out for Beta users, wider release coming soon. 🎥 Quick walkthrough from our product team below ⬇️ Got feedback or feature requests for the page? Drop them in the thread, we're actively iterating. 🚀

Webhook page_demo.mov

🎉 2

🔥 2

Dasha

05/27/2026, 3:18 PM

🚀 New API Endpoints: Inbox, Background Jobs & Email Validations Hey everyone 👋 We've just shipped a fresh batch of brand-new endpoints. Three entirely new sections in the API where there was nothing before: 📥 Inbox ⚙️ Background Jobs ✅ Email Validations Inbox status: most endpoints are live and ready to use. Two are still coming soon (we wanted to give you a heads-up so you can plan ahead): • Generate an AI draft reply • Submit feedback on a generated AI draft Background Jobs & Email Validations: all endpoints are live and ready to go. 🎉 Have a look, build something cool, and drop your feedback in the thread. Especially curious to hear what you're planning for the AI draft endpoints once they ship.

🎉 1

Dasha

06/08/2026, 5:41 PM

📢 Reply API V3 — Changelog (June 5) Hey folks! 👋 We've shipped a chunky batch of API updates - full AI SDR Sequences coverage, new Live Data + Intent Signals typeahead endpoints, and a few breaking changes worth a quick look before your next deploy. Quick rundown below 👇 ────────────────────────────── 🚨 Breaking changes — please check these 1️⃣ totalCount
removed from Contacts endpoints

Affects

GET /v3/contacts

POST /v3/contacts/filter

Response now returns only
items
+
hasMore
. If you read
totalCount
, switch to cursor-style pagination using
hasMore
.

2️⃣ Sequence Preview feedback path moved — now under the

ai-sdr

segment:

/v3/sequences/ai-sdr/{id}/contacts/{id}/preview/steps/{step_id}/feedback

3️⃣ Tag renames

AI SDR Live Data

→

Live Data

AI SDR Contact Enrichment

→

Contact Enrichment

JasonAi credits

→

Reply credits

────────────────────────────── ✨ AI SDR Sequences — now fully covered All 14 endpoints are live under

/v3/sequences/ai-sdr/{id}

(Create = base path, no

{id}

Copy code

Action                         Method          Path
Create sequence                POST            (base)
Get settings                   GET             /settings
Update settings                PATCH           /settings
Enable autopilot               POST            /autopilot/enable
Disable autopilot              POST            /autopilot/disable
Force-start autopilot          POST            /autopilot/force-start
Set approval mode              PUT             /approval-mode
Set generated step types       PUT             /generated-step-types
Get/Set/Remove playbook        GET/PUT/DELETE  /playbook
Get/Set/Remove knowledge base  GET/PUT/DELETE  /knowledge-base

Heads up on force-start errors:
409 autopilotDisabledByContactsLimit
(limit reached) ·
409 autopilotDisabledByLackOfContacts
(no contacts) ·
404
on playbook/KB DELETE when nothing's attached.

✨ More new endpoints • Intent Signals (typeahead) —

GET /v3/ai-sdr/intent-signals/industries

…/technologies

⚠️ These return Sales Navigator strings — different from Live Data filter-value industries (Reply taxonomy), so don't mix them up. • Live Data Filter Values are live (finally off "coming soon" 🎉) — 5 typeahead endpoints under `/v3/live-data/filter-values`: industries · departments · seniorities · locations · job-titles ────────────────────────────── 🔧 Smaller improvements • Inbox — filter by owner via new `ownerUserId`:

POST /v3/inbox/threads/filter

(body, integer nullable) and

GET /v3/inbox/threads/categories

(query param, integer). Target user must be on your team, else

. • Rate limiting — all

responses now include a

Retry-After

header (integer, seconds) so you can back off cleanly. • Sidebar reordered to strict A–Z, parent-first grouping. • Auth guide rewritten — now covers Master API key + Organization API key (with impersonation details). ────────────────────────────── 📚 Full docs: https://docs.reply.io/api-reference Questions or hit a snag migrating? Drop them in the thread 🧵

Dasha

06/15/2026, 5:12 PM

🚀 Reply API V3 — Changelog TL;DR: AI-drafted replies now surface in the Inbox API, companyDescription limits triple to 3,000 chars, and the AI SDR Pending Approvals API is fully documented ahead of its mid-June release. One breaking change — the AI voice LinkedIn endpoint drops playback-speed settings. 🟢 New: live now Inbox - AI Draft support GET /v3/inbox/threads/{id} now returns AI reply data: • aiDraft (nullable) - pending AI-drafted reply: message (HTML), cc, bcc, attachments. null if no active draft. • aiSentType (nullable) on email & LinkedIn messages - aiGenerated / aiGeneratedEdited / autoSent. Offers: larger company description • companyDescription max length raised 1,000 → 3,000 on create + patch. icp / reasonForOutreach stay 1,000. 🔜 Coming soon - targeted late-June 2026 AI SDR Pending Approvals - full API coverage • GET /v3/ai-sdr/approvals — list (filters sequenceId / userId / channel / sortMode, top/skip) • POST /v3/ai-sdr/approvals/send-bulk — batch of 1–20 pairs • GET|POST|DELETE /v3/sequences/ai-sdr/{sequence_id}/contacts/{contact_id}/approval[/send|/regenerate] • Regenerate is async: returns 202 {jobId} → poll GET /v3/background-jobs/{jobId} for subject/body. • Reject discards the draft and removes the contact from the sequence. AI SDR Sequences - limits & error contract • companyDescription 1,000 → 3,000 on offer-generation inputs. • Intent signals: technologyUsed & jobHiring now mutually exclusive. • Email-only sequences reject contacts without an email address. ⚠️ Breaking change

AI voice LinkedIn message - playback speed removed

Before: body accepted textToSpeechSettings with a speed multiplier.

After: textToSpeechSettings / speed removed; only text is accepted.

Action required: remove textToSpeechSettings from your request body - it has no effect.

📚 Full docs: https://docs.reply.io/api-reference

Dasha

06/30/2026, 3:45 PM

🚀 Reply API V3 — Changelog (June 30, 2026)

TL;DR: Mostly AI SDR + Live Data roadmap — search, autopilot-preview, and per-contact personalization endpoints are now fully documented ahead of their mid-June 2026 launch (still Coming Soon, not callable yet). The one live change is cosmetic: direct-outreach sends moved into their own Direct Outreach section.

🟢 Now Live Direct Outreach — sends get their own section. The 6 direct-send endpoints (already beta) moved out of Contacts into a new Direct Outreach group. Paths, methods, and behavior are unchanged — docs grouping only, no action needed. 🔜 Coming Soon (not callable yet) Live Data — Search API • Preview a search —

POST /v3/live-data/searches/preview

→

202 {jobId}

(no prospects, no credits) • Start a search —

POST /v3/live-data/searches

→

200 {searchId}

(note: 200, not 202; consumes credits) • List searches —

GET /v3/live-data/searches

(default

top

100) • Get a search —

GET /v3/live-data/searches/{id}

New

422 liveDataSearch.noMatchingContacts

. Status enum:

inProgress

finished

noMoreContacts

manuallyPaused

pausedDueToInsufficientCredits

. AI SDR — Autopilot preview •

POST /v3/sequences/ai-sdr/{sequence_id}/autopilot/preview

→

202 {jobId}

, poll

GET /v3/background-jobs/{id}

for

{contactsCount, companiesCount, sampleContacts}

. Try-before-save: filter config doesn't need saving first. AI SDR — Sequence preview & personalization • Get preview —

GET /v3/sequences/{sequence_id}/contacts/{contact_id}/preview

(generic, not AI-SDR-only) • Reset preview —

POST .../preview/reset

• Personalize —

POST /v3/sequences/ai-sdr/{sequence_id}/contacts/{contact_id}/preview/personalize

→

202 {jobId}

, progress

personalizedSteps/totalSteps

Early adopters: personalization endpoint renamed from
.../personalization/start
to `.../preview/personalize`; bulk stub removed. Never shipped, so no live impact.

🔧 Housekeeping • New Direct Outreach tag (after Custom Fields) • Live Data list + AI SDR Approvals list

top

default corrected 25 → 100 • Live Data search status enum casing fixed to camelCase •

totalSteps

documented in the personalization job result 📚 Full docs: docs.reply.io/api-reference

Dasha

07/08/2026, 3:51 PM

Hey everyone 👋 We've just released a new, extended version of our MCP! We've significantly expanded the functionality, so I'd recommend reconnecting to pick up all the updates. Enjoy, and please share your feedback, we'd love to hear your thoughts and ideas. All the details here: docs.reply.io/reply-mcp P.S. @Hessan Haiderzada, you've been especially interested in MCP, so tagging you directly, I think you'll want to give this a spin 🙂

🎉 1

❤️ 1

Dasha

07/17/2026, 11:17 AM

🚀 Reply API V3 — Changelog (July 16, 2026)

TL;DR: Sequence sorting & date filters,
emailProvider
on contact accounts, new team-users endpoint. Counts, unarchive, CSV export & email stats documented ahead of July releases.

🟢 Now Live • Workspace —

GET /v3/whoami/team-users

lists all users in your team:

teamId

teamName

userId

userName

userEmail

• Sequences —

GET /v3/sequences

gains

sortBy

(`created`/`name`/`status`),

sortDirection

(`asc`/`desc`, requires

sortBy

createdAfter

(ISO-8601); invalid sort →

400 sequence.invalidSort

• Contact Accounts — single-account responses add read-only

emailProvider

, resolved from the domain's MX records (

gSuite

office

gmail

, …);

null

when undetermined • AI SDR Reply Handlers (beta) — new

deliveryMode

(`draft`/`auto`/`stopAndNotify`) + `additionalNotificationEmail`;

stopAndNotify

is feature-gated →

403 knowledgeBase.stopAndNotifyNotEnabled

🔜 Coming Soon late July 2026: • Counts (return

{"count": N}

POST /v3/contacts/filter/count

•

GET /v3/sequences/count

•

GET /v3/sequences/{id}/contacts/count

• Unarchive:

POST /v3/sequences/{id}/unarchive

•

POST /v3/sequences/unarchive

(bulk, 1–100 ids, non-atomic) • Email account stats:

GET /v3/email-accounts/{id}/stats

•

POST /v3/email-accounts/stats

(

emailsSentToday

totalEmailsSent

dailySendingLimit

) mid-July 2026: •

GET /v3/holiday-calendars/{id}/export-csv

— CSV download, filename = calendar name 📚docs.reply.io/api-reference/introduction

Dasha

07/27/2026, 3:27 PM

🚀 Reply API V3 — Changelog (July 27, 2026) TL;DR: New beta endpoint for a contact's inbox threads, a new channel-agnostic

contact_replied

webhook,

email_replied

now fires on every email reply (was first-reply-only), richer

sequence_fields

on all sequence webhooks — and one breaking change:

email_replied

dropped its

reason

field. All webhook changes are v3 only — v2 is untouched. 🟢 Now Live Contacts — inbox threads on a contact's timeline (Beta) •

GET /v3/contacts/{id}/inbox-threads

— email/LinkedIn threads for a contact; paged

top

(default 25, max 1000) / `skip`; scope

contacts:read

. New Contact Inbox Thread schema. Inbox — reply to a specific message •

messageId

(string, nullable) now returned on email & LinkedIn messages. •

replyToMessageId

(string, nullable) on the email send request — omit to reply to the last message in the thread. Webhooks —

email_replied

now fires on every reply • Previously it fired only on the first reply from a contact, so any replies after that weren't tracked. Now you get an event for every new email reply. Webhooks — new

contact_replied

event • Channel-agnostic status signal — fires whenever a contact's status changes to "Replied", regardless of how (email, LinkedIn, SMS, or manual/API mark). `reason`:

EmailDetected

LinkedInDetected

SmsDetected

StatusSetManually

. • Payload:

reply_date

reply_message_id

reason

contact_fields

contact_custom_fields

sequence_fields

. Correlate to per-channel webhooks via

reply_message_id

. Webhooks — richer

sequence_fields

• Now consistently carries

id

name

step_number

across sequence events (was

id

-only or `id`+`step_number` on some). •

linkedin_reply_categorized

now carries

sequence_fields

+ opt-in

linkedin_message

(needs

includeLinkedInMessageText

). Auth — stable error codes for Organization API keys • Rejected impersonation →

application/problem+json

with a `code`:

403 USER_REQUIRED

401 USER_NOT_FOUND

403 TEAM_NOT_ACCESSIBLE

. Breaking Changes

email_replied

—

reason

removed; scope narrowed Before: fired only on the first detected reply or when contact status was set to "Replied"; payload had

reason

(`EmailDetected`/`StatusSetManually`). After: no `reason`; fires on every detected email reply — and only on detected replies. Manual "Replied" marks now flow through

contact_replied

. Action: stop reading `reason`; expect multiple events per contact; to catch manual marks subscribe to

contact_replied

(

reason: StatusSetManually

). Note: these webhook changes apply to v3 webhooks only — v2 behavior is unchanged. 📚 Full docs: docs.reply.io/api-reference

🎯 1

Dasha

08/11/2026, 1:16 PM

🚀 Reply API V3 — Changelog (Aug 11, 2026)

TL;DR: The AI SDR API surface is now live in Beta — 39 endpoints across AI SDR sequences, pending approvals, Live Data, sequence preview, offer generation, intent signals, the Strategist, and Inbox AI drafts moved from "coming soon" to Beta. No breaking changes.

🟢 Now Live (Beta) AI SDR Sequences — full lifecycle via API •

POST /v3/sequences/ai-sdr

— create an AI SDR sequence • `GET`/`PATCH /v3/sequences/ai-sdr/{sequence_id}/settings` — read / update AI settings •

POST …/autopilot/enable

…/disable

…/force-start

…/preview

•

PUT …/approval-mode

PUT …/generated-step-types

• `GET`/`PUT`/`DELETE …/playbook` and

…/knowledge-base

AI SDR Pending Approvals •

GET /v3/ai-sdr/approvals

— list pending approvals •

POST /v3/ai-sdr/approvals/send-bulk

— send up to 20 at once • `GET`/`DELETE /v3/sequences/ai-sdr/{sequence_id}/contacts/{contact_id}/approval` — inspect / reject •

POST …/approval/send

POST …/approval/regenerate

Live Data •

POST /v3/live-data/searches/preview

— free preview (async job) • `POST`/`GET /v3/live-data/searches` — start / list searches •

GET /v3/live-data/searches/{id}

GET /v3/live-data/filter-values/…

(typeaheads) Sequence Preview & AI Personalization •

GET /v3/sequences/{sequence_id}/contacts/{contact_id}/preview

POST …/preview/reset

•

POST /v3/sequences/ai-sdr/{sequence_id}/contacts/{contact_id}/preview/personalize

More now in Beta: offer generation (

POST /v3/ai-sdr/offers/generate

+ attachments), Strategist (

POST /v3/ai-sdr/strategist/start

), intent-signal typeaheads, Inbox AI drafts (

POST /v3/inbox/threads/{id}/ai-draft/generate

) 🔜 Coming Soon • Mid August 2026: duplicate offer / duplicate playbook, holiday-calendar CSV export • Late August 2026: AI Insights, AI Web Search, Contact Enrichment (5 endpoints), count endpoints (contacts/sequences), sequence unarchive, email-account stats, AI feedback endpoints 📝 Docs: restructured into API Reference / MCP / Skills / CLI tabs, new MCP + CLI (

reply-cli v0.3.0

) + Skills + agent-guide sections, 8 task guides, and a new Slack developer community. 📚 Full documentation

Dasha

08/27/2026, 12:01 PM

🚀 Reply API V3 — Changelog (August 27, 2026)

Three small additions and one correction. If you create schedules through the API, read the last part.

🟢 Now Live Inbox — reply to a different address.

POST /v3/inbox/threads/{id}/messages

takes an optional

to

(

string

format: email

, nullable), for when a colleague or a shared inbox answered instead of the contact. Omit it and nothing changes. Note: setting it switches off link and open tracking for that message. Settings — tracking and plain text.

GET

PATCH /v3/settings

now expose

isLinksTrackingEnabled

and

isSendingPlainTextEnabled

(both

boolean

, nullable) in the

emails

section. They don't mix — plain text can't be enabled while open or link tracking is on:

Copy code

400  settings.invalidOperation
     Cannot enable plain text mode together with open or click tracking

LinkedIn — delay between actions.

PUT /v3/linkedin-accounts/{id}/limits

takes

actionDelayMinSeconds

(

>= 30

, default

) and

actionDelayMaxSeconds

(<= 86400`, default

, with

max - min>

= 60`). Reply picks a random delay in

[min, max]

before each action. ⚠️ That

PUT

replaces the whole limits object — omit the two fields and you get the 60–120s defaults, which changes pacing even if your code doesn't change. ⚠️ Schedules —
timezoneId
never took IANA identifiers Our docs said IANA (

"America/New_York"

). The API takes Windows identifiers and rejects IANA. Nothing changed in the API — the documentation was wrong, and anyone who trusted it has been getting rejections. Map before calling

POST /v3/schedules

or `PUT /v3/schedules/{id}`:

"America/New_York"

→

"Eastern Standard Time"

"Europe/London"

→

"GMT Standard Time"

. 📝 Also: 276 endpoint descriptions across 30 areas were rewritten. No contract changes, but they now spell out behavior that was always true — notably on

PATCH /v3/settings

, where an explicit

null

clears a text field but is ignored on booleans and numbers, and

calls.resolutions

replaces the whole list, so a custom resolution you omit is deleted. 📚 Full documentation

🙌 2

Dasha

09/17/2026, 12:18 PM

🚀 Reply API V3 — Changelog (September 17, 2026)

Six AI Prompts endpoints, holiday-calendar CSV export and the AI SDR duplicate operations came out of Coming soon. Contact import gains a
contactId
dedup key, sequence contacts can be repointed at a different sending mailbox, and Mailbox Procurement is documented ahead of a late-September release.

🟢 Now Live AI Prompts — all six operations out of Coming soon: •

GET /v3/ai-prompts

— List AI prompts •

POST /v3/ai-prompts

— Create a personal AI prompt •

PUT /v3/ai-prompts/{id}

— Update a personal AI prompt •

DELETE /v3/ai-prompts/{id}

— Delete a personal AI prompt •

POST /v3/ai-prompts/preview/sample

— Preview a comment prompt against a sample post •

POST /v3/ai-prompts/preview/linkedin-post

— Preview a comment prompt against a LinkedIn post Scopes unchanged:

sequences:read

to list,

sequences:write

to create/update/delete,

sequences:operate

for both previews. Sequence Contacts — repoint contacts at a different sending mailbox: •

POST /v3/sequences/{id}/contacts/set-email-account

(

sequences:operate

) Body takes

contactIds

(1–100, positive, no duplicates) and

emailAccountId

. The mailbox must already be linked to the sequence — link it with

POST /v3/sequences/{id}/email-account-links

first, otherwise you get

400 sequenceContact.invalidInput

. The

is non-atomic:

assigned

lists the ids now sending from that mailbox (including ones already on it, so repeating the call is safe),

notProcessed

maps the rest to a reason (

notInSequence

unknown

). Owner-only (

); archived sequence →

409 sequenceContact.sequenceArchived

. Also out of Coming soon: •

GET /v3/holiday-calendars/{id}/export-csv

— Export a holiday calendar as CSV •

POST /v3/ai-sdr/offers/{id}/duplicate

— Duplicate an offer •

POST /v3/ai-sdr/playbooks/{id}/duplicate

— Duplicate a playbook Contact import — match existing contacts by Reply contact id

POST /v3/contacts/import

accepts

options.keys.contactId

(boolean, default

false

) and a

contactId

on each item (integer, nullable). Two rules worth noting: the key is not additive — when enabled it decides the match on its own and the other keys are not consulted; and every item must then carry a

contactId

, where an id that doesn't resolve inside your team is imported as a new contact rather than failing. Available only to teams it has been enabled for — otherwise

403 contact.forbidden

with

Matching contacts by contactId is not enabled for your team.

🔜 Coming Soon Mailbox Procurement — buy domains and mailboxes through the API (targeted: late September 2026, not callable yet): •

GET /v3/email-accounts/procurement/domains/available

(

channels:read

) •

POST /v3/email-accounts/procurement/domains/estimate

(

channels:read

) •

POST /v3/email-accounts/procurement/domains/purchase

(

channels:write

) •

POST /v3/email-accounts/procurement/users

(

channels:write

) What to expect: every monetary amount is in cents (

= $12.99). Domain search takes one label (up to 63 chars, letters/digits/hyphens, no leading or trailing hyphen) against up to 5 TLDs from

com

net

org

biz

live

. Orders specify

serviceProvider

google

microsoft

and 1–5 mailboxes, with unique usernames and exactly one

isAdmin: true

. Purchase is async —

jobId

, poll

GET /v3/background-jobs/{jobId}

. Registration is a precondition: every other procurement endpoint returns

until

POST /v3/email-accounts/procurement/users

succeeds. Contacts — bulk opt-out (targeted: late September 2026): •

POST /v3/contacts/set-opted-out

(

contacts:operate

) —

contactIds

(1–100, no duplicates) +

isOptedOut

. Opting out finishes the contacts in every sequence they're in; the

contact_opted_out

webhook fires once per enrolled sequence, and not at all for a contact in no sequence. The

lists only per-item failures (

notFound

invalidInput

);

{}

means all succeeded. ⚠️ Breaking Changes 1.
steps
is now required when creating a sequence

Before:
POST /v3/sequences
required only
name
.

After:
steps
is required with
minItems: 1
— a sequence with no steps has nothing to send and cannot be started.

Action required: if you create an empty sequence and add steps in a follow-up call, move at least one step into the create request body.

2. Inbox category pages moved

Before:
/api-reference/inbox/list-inbox-categories
and six sibling pages.

After: the
Inbox
tag split into Inbox (Threads, 13 ops) and Inbox Categories (7 ops); category pages now live at
/api-reference/inbox-categories/…
.

Action required: documentation links only — no endpoint, request or response changed. Redirects are in place, so existing links still resolve.

Housekeeping • The per-endpoint Beta banner is replaced by a New banner across 57 operations, meaning "recently shipped" with an invitation to give feedback — never a breaking-change warning. The API-wide Beta signal now appears once, as a pill beside the version. • The API reference is regrouped into 12 top-level sections, including the new Mailbox Procurement. • New schema pages: Available Domain, Estimate Seats. • The spec now declares

openapi: 3.2.0

(was

3.1.0

) to use hierarchical tags. 📚 Full documentation

Andrii Samokysh

10/23/2026, 6:53 AM

Now Live October 22 • Webhooks — `email_replied`: new

email_to

email_cc

email_bcc

(arrays of addresses from the message headers). Empty headers are omitted, so

email_bcc

is absent in practice. Not gated by

includeEmailUrl

includeEmailText

. • `POST /v3/sequences`: every

settings

field is now optional and nullable; an omitted field takes its documented default (e.g.

daysToFinishProspect

emailSendingDelaySeconds

dailyThrottling

disableOpensTracking

true

repliesHandlingType

markAsFinished

generatedTaskOwner

`sequence`;

emailsCountPerDay

= your workspace default). • `POST /v3/contacts/import`:

options.keys.contactId

(boolean, default

false

) +

contactId

per item. The key is not additive, every item must then carry a

contactId

, and an id that doesn’t resolve inside your team imports a new contact. • Inbox: new message status code

linkedInInvalidUrl

(LinkedIn profile URL missing/invalid or contact deleted; resending fails until the URL is fixed). • AI SDR: unknown personalization variables now return a descriptive

—

offer.invalidParameter

(create/patch),

playbook.invalidParameter

(create/patch),

salesAgentSequence.invalidParameter

(settings patch). Example:

Text contains unknown variables. Unknown variables: {{Your_Company}}. Fields: callToActions.

October 8 • Billing:

GET /v3/billing

(

billing:read

) is out of Coming soon. • LinkedIn Accounts:

PUT

DELETE /v3/linkedin-accounts/{id}/voice

(

channels:write

) are out of Coming soon. Unlinking falls back to the team’s default voice. • Accounts — custom field values:

customFields[]

on account responses, create and update (by

id

name

value

≤ 4000).

PUT

writes only the values you send; bad value →

400 contactAccount.invalidInput

, feature off → `403 contactAccount.featureDisabled`; on bulk create only the failing item fails (error

). • Custom Fields: new

entityType

(

contact

account

) on the model, create (default

contact

), update (must match) and

GET /v3/custom-fields?entityType=

. • Contacts:

stage

(name, nullable) on read,

PATCH

and import.

null

is the only value that clears a stage; unknown/blank name →

listing available names. • `GET /v3/whoami/team-users`: new

isOwner

teamRole

organizationRole

. • Sequences:

GET /v3/sequences/{id}/steps

is ordered by position (empty array when none,

404 sequenceStep.notFound

);

POST /v3/sequences

documents

sequence.invalidStep

sequence.emptySubject

. • Reporting / sequence stats: plain-date

to

covers the whole day, UTC-timestamp

to

is an exclusive bound, times are read in the user’s time zone (the

suffix does not shift them). • Rate limits: weekly quota documented — 100/min, 3,000/h, 150,000/week per user. Prompt Actions: up to 6

outputs

(was 5). 🔜 Coming Soon AI SDR Sequence Structures (targeted: late October 2026) •

GET /v3/ai-sdr/sequence-structures

GET /v3/ai-sdr/sequence-structures/{id}

(

ai-sdr:read

) •

GET

PUT /v3/sequences/ai-sdr/{sequence_id}/structure

POST /v3/sequences/ai-sdr/{sequence_id}/structure/save-as

(

ai-sdr:write

for

PUT

save-as

) Ids are

g-{number}

(library) or

c-{number}

(user-saved).

PUT

replaces the sequence’s steps and returns

409 sequenceStructure.changeBlocked

once messages were sent.

POST /v3/sequences/ai-sdr

takes an optional `structureId`;

502 salesAgentSequence.structureNotSuggested

means the sequence was created without steps (id in

detail

). Teams (targeted: early November 2026) Notes API (late October 2026):

GET /v3/contacts/{id}/notes

GET /v3/contacts/{id}/notes/{note_id}

GET

POST /v3/contact-accounts/{id}/notes

GET /v3/contact-accounts/{id}/notes/{note_id}

. AI SDR playbooks (late October 2026): playbook draft endpoints under

/v3/sequences/ai-sdr/{sequence_id}/playbook-draft

(

generate

regenerate

example-files

save

POST /v3/ai-sdr/playbooks/{id}/regenerate

GET /v3/ai-sdr/playbooks/{id}/sequences

. Drafts are deleted after 1 hour of inactivity. Mailbox Procurement (late October 2026 — moved from late September): new

GET /v3/email-accounts/procurement/domains

POST .../domains/{domainId}/mailboxes/estimate

POST .../domains/{domainId}/mailboxes/purchase

GET .../procurement/geo/states

. Rate limits (mid October 2026):

GET /v3/rate-limits

— your usage per limit and window. ⚠️ Breaking Changes 1. Unknown or malformed template variables are rejected with
400
(Oct 8) > Before: variable names were documented as case-sensitive; no rejection of unknown names was documented. > After: email template create/update,

POST /v3/sequences

and step/variant create/update/bulk (email and SMS steps) return

, e.g.

Unknown variable {{firstName}}. Did you mean {{FirstName}}? …

. LinkedIn, WhatsApp, call and task steps are not checked. > Action required: check variable names against

GET /v3/email-templates/variables

and use the exact casing. 2. AI SDR sequence steps now come from sequence structures (Oct 22, with the late-October structures release) > Before:

POST /v3/sequences/ai-sdr

required `sequenceStructureSetup`;

PUT /v3/sequences/ai-sdr/{sequence_id}/generated-step-types

set the channels; settings returned

generatedStepTypes

. > After:

sequenceStructureSetup

is removed and

structureId

is optional;

PUT .../generated-step-types

and

generatedStepTypes

are removed. > Action required: stop sending

sequenceStructureSetup

(send

structureId

or nothing), move to

PUT .../structure

, stop reading

generatedStepTypes

. Housekeeping • LinkedIn account limits:

limitsMode

is documented as

fixed

ranged

(the casing

GET

returns). • New schema pages (Sequence structure, Contact Note, Playbook draft, Rate Limits and more); new sidebar groups for Sequence Structures, Teams, Purchased domains, Mailboxes on existing domains and Rate Limits. • Authentication guide: new

teams

scope row; MCP pages updated for the weekly limit and

Retry-After

. 📚 Full documentation

🙌 1

Andrii Samokysh

10/23/2026, 1:38 PM

New API endpoint: Add existing users to a team (in bulk) New endpoint

POST /v3/teams/{team_id}/users/bulk

adds up to 50 users per call, each with a role of that team