> ## Documentation Index > Fetch the complete documentation index at: https://docs.sendler.ai/llms.txt > Use this file to discover all available pages before exploring further. # Send Connection Request > Request a LinkedIn connection from a workspace sender. Requires `inbox:send`. Supply `senderId` and `recipientLinkedinUrl`. The optional note field is **`message`**, at most 300 characters; MCP calls the corresponding argument `note`. Sender account capability and daily connection limits still apply. Use an [Idempotency-Key](https://docs.sendler.ai/api-reference/behavior.md#idempotent-writes) and inspect uncertain outcomes before retrying. A successful REST request executes directly, without an MCP confirmation token. ## OpenAPI ````yaml POST /v1/inbox/connect openapi: 3.0.0 info: title: SendPilot External API description: > # SendPilot External API v1 Public API for integrating with SendPilot programmatically. ## Authentication Requests authenticate with a workspace API key in the `X-API-Key` header: ``` X-API-Key: sp_live_your_api_key_here ``` `Authorization: Bearer sp_live_...` (or `sp_test_...`) is also accepted. Prefer one header; `X-API-Key` takes precedence. Clerk OAuth tokens for MCP connection URLs are not public REST API keys. A key only reaches what its scopes and creator's current workspace permissions allow. Each operation lists the scopes it needs in `x-sp-scopes`. Use `GET /v1/me` to inspect the effective permissions and subscription state. ## Rate Limits - **Per API Key**: 300 requests/minute - **Per Workspace**: 50,000 requests/day Responses that reach the rate limiter include these headers; early authentication failures may not: - `X-RateLimit-Limit`: Max requests per minute - `X-RateLimit-Remaining`: Remaining requests in current window - `X-RateLimit-Reset`: Epoch timestamp when limit resets ## LinkedIn Sender Limits When using messaging endpoints, these headers show account limits: - `X-LinkedIn-Messages-Remaining`: Messages left today - `X-LinkedIn-Connections-Remaining`: Connection requests left today - `X-LinkedIn-Quota-Reset-At`: When daily limits reset ## Retries Send an `Idempotency-Key` header on POST and PATCH requests. A retry with the same key within 24 hours returns the first result (with `Idempotent-Replayed: true`) instead of acting twice. The same key with a different request answers 422 `IDEMPOTENCY_KEY_REUSED`; a retry while the first request still runs answers 409 `IDEMPOTENCY_IN_PROGRESS`. Once execution is attempted, its response (including errors) is retained. If a process stops or result storage fails after the attempt, the durable record blocks another execution for the replay window and returns 409 `IDEMPOTENCY_IN_PROGRESS` or `IDEMPOTENCY_REPLAY_UNAVAILABLE`. Check the operation's status before using a new key: an uncertain outcome may already have acted. Requests refused before execution is claimed, such as authentication or rate-limit refusals, can be retried with the same key. ## Webhooks Configure webhooks in the SendPilot app to receive real-time events: - `lead.tag.updated` - `lead.updated` - `message.sent` - `reply.received` - `connection_request.sent` - `connection_request.accepted` - `campaign.started` - `campaign.paused` - `campaign.resumed` - `campaign.finished` ## Error Codes Errors return a JSON body. Use the machine-readable `code` when supplied; validation messages can be arrays, and some legacy errors omit `code`. A typical error is: ```json { "statusCode": 400, "error": "Bad Request", "code": "SPECIFIC_ERROR_CODE", "message": "Human-readable description" } ``` version: '1.0' contact: name: SendPilot Support url: https://sendpilot.ai email: support@sendpilot.ai servers: - url: https://api.sendpilot.ai security: [] tags: [] paths: /v1/inbox/connect: post: tags: - External API - Inbox summary: Send LinkedIn connection request description: >- Sends a connection request to a LinkedIn profile. Optional connection note only works for premium accounts. operationId: inbox.sendConnectionRequest parameters: [] requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/SendConnectionRequestDto' responses: '200': description: Connection request sent content: application/json: schema: $ref: '#/components/schemas/SendConnectionRequestResponseDto' '400': description: Invalid request or sender not found '429': description: Daily connection limit exceeded security: - api-key: [] components: schemas: SendConnectionRequestDto: type: object properties: senderId: type: string description: LinkedIn sender ID to use (must be connected to workspace) example: sender_123 recipientLinkedinUrl: type: string description: LinkedIn URL of the recipient to connect with example: https://www.linkedin.com/in/johndoe/ message: type: string description: Optional connection note (only works for premium LinkedIn accounts) example: Hi John, I would love to connect with you! maxLength: 300 required: - senderId - recipientLinkedinUrl SendConnectionRequestResponseDto: type: object properties: success: type: boolean description: Success status example: true requestId: type: string description: Request ID example: conn_123 recipientLinkedinUrl: type: string description: Recipient LinkedIn URL status: type: string description: Connection status example: sent enum: - sent - already_connected - failed error: type: string description: Error message if failed timestamp: format: date-time type: string description: Timestamp when request was sent required: - success - requestId - recipientLinkedinUrl - status - timestamp securitySchemes: api-key: type: apiKey in: header name: X-API-Key description: 'API key for authentication (prefix: sp_live_ or sp_test_)' ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.