Create an AI SDR sequence
Creates a new AI SDR sequence with its full AI SDR settings payload. The sequence is seeded either from a saved offer (offerId) or from inline offer-context inputs (offerGenerationInputs) — exactly one is required.
Sequence steps are generated by the AI SDR — clients do not pass steps here. Generic sequence fields (name, scheduleId, settings, emailAccounts, linkedInAccounts) are accepted and applied after the AI SDR-specific setup.
Requires the AI SDR feature on the caller’s team.
Authorizations
Authenticate every request with a Bearer token. Pass your Reply API key in the
Authorization header:
Authorization: Bearer <your-api-key>
Get your API key from the Reply dashboard: Settings → API Key.
Body
Create a new AI SDR sequence with its full AI SDR settings. Generic sequence fields (name, scheduleId, settings, emailAccounts, linkedInAccounts) are accepted here too — sequence steps are NOT, because the AI SDR generates steps automatically.
Exactly one of offerId or offerGenerationInputs must be provided.
At most one intent signal per type is allowed.
Display name of the sequence
1 - 255Shape of the sequence steps the AI SDR should generate on create.
Id of the send schedule to use
x >= 1Generic sequence settings (throttling, tracking, replies handling). Optional — sensible defaults are applied when omitted.
Email account ids to link to the sequence
x >= 1LinkedIn account ids to link to the sequence
x >= 1Id of an existing offer to seed the sequence from. Mutually exclusive with offerGenerationInputs.
x >= 1Inline offer-context inputs. Mutually exclusive with offerId.
Tone-of-voice configuration. Omitted fields fall back to defaults (English (US), Confident, Medium, Man).
LLM provider that generates the outreach content. Defaults to claude when omitted.
claude, openAi, gemini, mistral Research signals the AI SDR is allowed to use in personalized messages
Autopilot configuration applied immediately after sequence creation
Composite playbook id ({prefix}-{numericId}) to attach to the sequence. Prefix encodes the playbook scope: g for global, o for organization, t for team (e.g. g-1, o-42, t-7).
1Id of a knowledge base to connect to the sequence
x >= 1Initial approval-mode setting. Cannot be changed via the AI SDR sequence PATCH — use PUT .../approval-mode after creation.
Intent signals to attach. At most one per type.
Response
AI SDR sequence created
AI SDR sequence creation response — extends the generic sequence response with the AI SDR-specific settings under aiSdr.
Unique identifier for the sequence
Identifier of the user who owns the sequence
Name of the sequence
Sequence creation timestamp with timezone offset
Current status of the sequence
new, active, paused Indicates if the sequence is archived
Overall health status of the sequence. Indicates whether the sequence can operate normally or has issues that need attention.
healthy— Sequence is functioning normally with no issuesstalled— Sequence has stalled and is not progressingdegraded— Sequence is running but with reduced effectivenessblocked— Sequence cannot proceed due to critical issues
healthy, stalled, degraded, blocked Schedule ID
Email accounts used to send emails for this sequence
LinkedIn accounts linked to this sequence
Settings configuration for a sequence
Array of sequence steps
Email step with variant configuration
- Email
- LinkedIn Message
- LinkedIn Connect
- LinkedIn InMail
- LinkedIn View Profile
- LinkedIn Endorse Skills
- LinkedIn Voice Message
- LinkedIn Like Recent Posts
- LinkedIn Follow Profile
- Call
- SMS
- WhatsApp
- Zapier
- Task
- Condition
AI SDR-specific settings of a sequence. Returned by GET .../settings and PATCH .../settings. Generic sequence fields (name, schedule, accounts, etc.) come from the generic GET /v3/sequences/{id} and are NOT included here.