> ## Documentation Index > Fetch the complete documentation index at: https://bavlio.com/docs/llms.txt > Use this file to discover all available pages before exploring further. # Reply To Conversation > Reply into an existing LinkedIn chat from the inbox. The other LinkedIn write endpoints on this router carry ``_require_non_production`` because they were built as a playground: they take an ``account_id`` from the client and can open a conversation with anyone. This one is deliberately production-safe and narrower on every axis that matters: * the sending account is resolved SERVER-SIDE from the caller's org, so a client cannot nominate someone else's connected account; * the chat must be bound to that org and account: either a local ``LinkedInMessageStore`` row, or (only with an explicitly selected org-owned account) the exact API chat verified fresh against the provider as an ordinary one-to-one chat. A foreign, guessed or mismatched ``chat_id`` is a 404, not a send; * it can only continue a conversation that already exists. There is no path here to start one, which is what makes it safe without the daily/hourly connection caps the playground endpoints need. It mirrors ``list_conversation_messages`` above — same org dependency, same account resolution, same store binding — because reading and replying to a thread should be authorised identically. ## OpenAPI ````yaml https://api.bavlio.com/openapi.json post /api/v1/linkedin/conversations/{chat_id}/reply openapi: 3.1.0 info: title: Bavlio API description: >- Outreach automation API — lead enrichment, email campaigns, LinkedIn automation, and AI agent. Full docs at https://docs.bavlio.com. version: 1.0.0 x-guidance: product: Bavlio Data API protocols: - x402 description: >- Pay-per-call B2B data endpoints for AI agents. USDC on Base mainnet via x402 v2. See /openapi.json for pricing. Landing page + marketing details: https://bavlio.com/developers network: eip155:8453 docs_url: https://docs.bavlio.com x-payment-merchant-address: '0xcca4725e0e931982830bff523b141e1a7a58d3e7' servers: [] security: [] tags: - name: API Keys description: >- Create, list, and revoke API keys for programmatic access. JWT authentication required. - name: credits description: Credit balance, cost estimates, and transaction history. - name: campaigns description: Create, manage, launch, and monitor email campaigns. - name: campaign-personalization description: AI-powered email subject and body personalization for campaigns. - name: email-finder description: Find and verify email addresses for prospects. - name: sender-profiles description: Manage sender identity profiles for outreach. - name: sender-profile-chat description: Interactive chat for building sender profiles. - name: templates description: Email template CRUD. - name: suppression description: Do-not-contact suppression list management. - name: datasets description: Lead dataset listing, download, and updates. - name: user-extractions description: Uploaded lead file management and processing. - name: enrichment-chat description: Interactive chat for data enrichment sessions. - name: enrichment-processing description: Batch lead enrichment processing. - name: File Upload (CSV/Excel) description: Upload and preview CSV/Excel lead files. - name: connected-accounts description: LinkedIn account connections via Unipile. - name: linkedin-content description: LinkedIn post creation, scheduling, and management. - name: linkedin-playground description: Direct LinkedIn actions (profile lookup, messaging, engagement). - name: bavimail-proxy description: BaviMail domain, alias, and email management. - name: agent description: Bavi AI agent chat with SSE streaming. - name: notifications description: Notification preferences and delivery. - name: referral description: Referral code management and stats. - name: feedback description: User feedback submission. - name: authentication description: JWT token verification and user info. - name: users description: User quota and usage information. - name: Health description: Health checks and system status. - name: Personalization description: Core lead personalization engine. - name: Session Management description: Enrichment session lifecycle. - name: Session Processing description: Session data processing. - name: webhooks description: Inbound webhook handlers (Stripe, BaviMail, Unipile). paths: /api/v1/linkedin/conversations/{chat_id}/reply: post: tags: - linkedin-playground summary: Reply To Conversation description: >- Reply into an existing LinkedIn chat from the inbox. The other LinkedIn write endpoints on this router carry ``_require_non_production`` because they were built as a playground: they take an ``account_id`` from the client and can open a conversation with anyone. This one is deliberately production-safe and narrower on every axis that matters: * the sending account is resolved SERVER-SIDE from the caller's org, so a client cannot nominate someone else's connected account; * the chat must be bound to that org and account: either a local ``LinkedInMessageStore`` row, or (only with an explicitly selected org-owned account) the exact API chat verified fresh against the provider as an ordinary one-to-one chat. A foreign, guessed or mismatched ``chat_id`` is a 404, not a send; * it can only continue a conversation that already exists. There is no path here to start one, which is what makes it safe without the daily/hourly connection caps the playground endpoints need. It mirrors ``list_conversation_messages`` above — same org dependency, same account resolution, same store binding — because reading and replying to a thread should be authorised identically. operationId: reply_to_conversation_api_v1_linkedin_conversations__chat_id__reply_post parameters: - name: chat_id in: path required: true schema: type: string minLength: 1 maxLength: 512 title: Chat Id - name: account_id in: query required: false schema: anyOf: - type: string format: uuid - type: 'null' title: Account Id - name: x-idempotency-key in: header required: false schema: anyOf: - type: string - type: 'null' title: X-Idempotency-Key requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/LinkedInConversationReplyRequest' responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/LinkedInActionResponse' '401': description: Unauthorized '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - HTTPBearer: [] components: schemas: LinkedInConversationReplyRequest: properties: text: type: string maxLength: 8000 minLength: 1 title: Text type: object required: - text title: LinkedInConversationReplyRequest description: >- Reply into an EXISTING LinkedIn chat from the inbox. Deliberately narrower than ``LinkedInMessageRequest``: no ``account_id``, no ``provider_id``, no ``identifier``. The account is resolved server-side from the caller's org and the chat is authorised against it, so a client cannot name the sending account or open a new conversation with someone who has not written to us. Replying to an existing thread is the one LinkedIn write that is safe to expose in production. LinkedInActionResponse: properties: success: type: boolean title: Success data: anyOf: - additionalProperties: true type: object - type: 'null' title: Data error: anyOf: - type: string - type: 'null' title: Error type: object required: - success title: LinkedInActionResponse HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type type: object required: - loc - msg - type title: ValidationError securitySchemes: HTTPBearer: type: http scheme: bearer ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.