> ## Documentation Index > Fetch the complete documentation index at: https://docs.erstan.com/llms.txt > Use this file to discover all available pages before exploring further. # Send a follow-up turn to a terminal run > Sends the next user message into a terminal run's conversation — the API twin of typing another message in the Erstan chat UI. Always poll the `runId` **returned by this call**: a completed chat-lane run of an interactive agent usually continues under the same `runId` with full conversation context, while failed/cancelled chat runs fork a fresh execution on the same thread, and agents that use post-run continuation (non-interactive agents and graphs with dedicated human-input nodes) fork the conversational post-run shell — including follow-up questions about completed structured runs. Only the same API key that started the run may continue it. Waiting runs keep their exact interaction contract (`/reply`, `/approval`); a run that is still executing returns `run_still_active`, and Builder Preview tests and structured runs of interactive agents return `run_not_continuable`. ## OpenAPI ````yaml /openapi/public-api.yaml post /v1/public/runs/{runId}/continue openapi: 3.1.0 info: title: Erstan Public Agent API version: 1.2.0 description: >- Start published Erstan agents from external systems and poll for results. API keys carry scopes (`agents:list`, `agents:read`, `agents:write`, `agents:run`, `runs:read`, `tasks:execute`, `documents:read`, `documents:write`, `files:read`, `files:write`). Agent access follows the key creator's current workspace and team permissions. Project and team allowlists (`allowedProjectIds`, `allowedTeamIds`) further restrict task and document and file access. Agent draft authoring and task/document/file tools are exposed through the MCP server at `/v1/mcp` rather than REST routes. servers: - url: https://api.erstan.com description: Public endpoint security: - PublicApiKey: [] tags: - name: Agents description: >- Discover and run published agents available to the key creator under their current permissions. - name: Runs description: Read run status and results. - name: MCP description: >- Model Context Protocol server for external clients. Agent Builder tools use `agents:read`, `agents:write`, or both, plus the key creator's live Erstan permissions. Task and content tools also apply project/team allowlists. paths: /v1/public/runs/{runId}/continue: post: tags: - Runs summary: Send a follow-up turn to a terminal run description: >- Sends the next user message into a terminal run's conversation — the API twin of typing another message in the Erstan chat UI. Always poll the `runId` **returned by this call**: a completed chat-lane run of an interactive agent usually continues under the same `runId` with full conversation context, while failed/cancelled chat runs fork a fresh execution on the same thread, and agents that use post-run continuation (non-interactive agents and graphs with dedicated human-input nodes) fork the conversational post-run shell — including follow-up questions about completed structured runs. Only the same API key that started the run may continue it. Waiting runs keep their exact interaction contract (`/reply`, `/approval`); a run that is still executing returns `run_still_active`, and Builder Preview tests and structured runs of interactive agents return `run_not_continuable`. operationId: continuePublicAgentRun parameters: - name: runId in: path required: true description: >- Terminal run ID returned by `POST /v1/public/agents/{agentId}/runs` or a previous continue call. schema: type: string format: uuid requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/RunContinueRequest' examples: followUp: value: message: Approve the proposed change set and create the sales orders. idempotencyKey: quote-chat-98342-turn-2 withAttachment: value: message: Here is the corrected workbook; re-run the checks. attachments: - name: corrected-orders.xlsx type: >- application/vnd.openxmlformats-officedocument.spreadsheetml.sheet base64: UEsDBBQAAAAI... idempotencyKey: quote-chat-98342-turn-3 responses: '200': description: Existing accepted turn returned for a matching idempotency key. content: application/json: schema: $ref: '#/components/schemas/RunContinueResponse' '202': description: >- Follow-up turn accepted. `runId` in the response is authoritative for polling — it matches the continued run for same-run continuations and is a new id when the turn forked a post-run or thread continuation. content: application/json: schema: $ref: '#/components/schemas/RunContinueResponse' '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '403': $ref: '#/components/responses/Forbidden' '404': $ref: '#/components/responses/NotFound' '409': description: >- The run is not a completed chat run ready for a follow-up turn, or another turn was accepted concurrently. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '503': description: >- The turn was valid but could not be queued; the completed state is unchanged and the caller may retry. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' security: - PublicApiKey: [] components: schemas: RunContinueRequest: type: object additionalProperties: false required: - message description: >- Request body for sending a follow-up conversation turn to a terminal run. properties: message: type: string minLength: 1 maxLength: 20000 description: Next user message for the conversation. attachments: type: array description: >- Optional attachments supplied with this turn. Each item needs `name`, `type`, and either `base64` or `storageKey`. items: $ref: '#/components/schemas/Attachment' idempotencyKey: type: string maxLength: 200 description: Reuse to prevent duplicate turns on retries of the same follow-up. RunContinueResponse: type: object additionalProperties: false required: - runId - agentId - laneId - status properties: runId: type: string format: uuid description: >- Run to poll for this turn — the same id for same-run continuations, a new id when the turn forked. agentId: type: string laneId: type: string nullable: true status: type: string description: '`running` once the follow-up turn is queued.' ErrorResponse: type: object required: - error properties: error: type: object required: - code - message properties: code: type: string message: type: string details: type: array items: $ref: '#/components/schemas/ErrorDetail' requestId: type: string Attachment: type: object additionalProperties: false required: - name - type properties: name: type: string type: type: string description: Content type, such as `application/pdf`. base64: type: string description: Base64-encoded file content. storageKey: type: string description: Existing Erstan storage key, when already uploaded. anyOf: - required: - base64 - required: - storageKey ErrorDetail: type: object required: - code - message properties: field: type: string code: type: string message: type: string responses: BadRequest: description: Request validation failed. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' Unauthorized: description: Missing, invalid, expired, revoked, or inactive API key. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' Forbidden: description: >- The API key lacks scope or the creator lacks current permission for the requested operation. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' NotFound: description: The requested agent or run was not found. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' securitySchemes: PublicApiKey: type: http scheme: bearer bearerFormat: ers_live API key description: >- Workspace API key created in Erstan Settings. Keys carry scopes (`agents:list`, `agents:read`, `agents:write`, `agents:run`, `runs:read`, `tasks:execute`, `documents:read`, `documents:write`, `files:read`, `files:write`). Agent access follows the key creator's current workspace, team, and agent permissions. Agent draft creation requires `agents:write`; update, validation, and publishing require both `agents:read` and `agents:write`; draft testing also requires `agents:run`. Keys with `tasks:execute` must include at least one allowed project or team (`allowedProjectIds` / `allowedTeamIds`); keys with document or file scopes must include at least one allowed team. Empty allowlists mean no access — the API fails closed. ````