> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.middesk.com/api-reference/agents/runs/list-agent-runs/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.middesk.com/_mcp/server. # List agent runs GET https://api.middesk.com/v1/runs Returns a paginated list of agent runs Reference: https://docs.middesk.com/api-reference/agents/runs/list-agent-runs ## Authentication - `Authorization` header (bearer token, required) — Bearer authentication of the form `Bearer `, where token is your auth token. ## Request ### Query parameters - `page` (integer, optional) — Page number - `per_page` (integer, optional) — Items per page ## Response ### 200 agent runs list with pagination - `object` (string, required) - `has_more` (boolean, required) - `data` (list of AgentsRun, optional) - `total_count` (integer, optional) ## Types ### AgentsRun A single agent run execution within an agent thread. - `object` (string, required) — Object type identifier. - `id` (string, required) — Unique identifier for the agent run. - `created_at` (datetime, required) — Timestamp when the agent run record was created. - `updated_at` (datetime, required) — Timestamp when the agent run record was last updated. - `status` (enum, required) — Current lifecycle status of the agent run. - Allowed values: `running`, `completed`, `error`, `interrupted` - `thread_id` (string, optional) — Identifier of the agent thread this run belongs to. - `business_id` (string, optional, nullable) — Middesk business the run is scoped to. Null when the run was started with an inline `business` payload whose business did not exist yet. - `agent` (string, optional) — Agent type that produced this run (for example, `cip_orchestrator`). - `requester` (AgentsRunRequester, optional, nullable) — Principal that initiated the run. Populated with the authenticated user's identity when a session caller initiates the run, and with the account when an API key caller initiates the run. `null` for system-initiated runs. - `started_at` (datetime, optional, nullable) — Timestamp when the agent run started. - `ended_at` (datetime, optional, nullable) — Timestamp when the agent run finished. - `output` (AgentsRunOutput, optional, nullable) — Artifact envelope produced by the agent. Non-null only when `status` is `completed`; `null` for `running`, `interrupted`, and `error` runs. - `interrupts` (list of AgentsRunInterruptsItem, optional) — Pending interrupt requests awaiting input. Empty when the run is not interrupted. To resume a run, POST the interrupt's `id` along with the caller-supplied input to `POST /v1/runs/{id}/resume`. - `steps` (list of AgentsStep, optional) — Top-level execution steps for this run. Opt-in: pass `?include=steps` on `GET /v1/runs/{id}` (or `POST /v1/runs/{id}/resume`) to populate this field. Omitted by default to avoid the cost of fetching step state from the underlying execution backend. ### AgentsRunRequester Principal that initiated the run. Populated with the authenticated user's identity when a session caller initiates the run, and with the account when an API key caller initiates the run. `null` for system-initiated runs. - `id` (string, required) — Identifier of the requesting user or account. - `name` (string, required) — Display name of the requester. - `email` (string, optional, nullable) — Email address of the requester. Populated for user requesters and `null` for account requesters. ### AgentsRunOutput Artifact envelope produced by the agent. Non-null only when `status` is `completed`; `null` for `running`, `interrupted`, and `error` runs. - `artifacts` (list of AgentsArtifact, required) — Artifacts produced by the agent during this run. ### AgentsRunInterruptsItem - `id` (string, required) — Identifier of the interrupt request. - `value` (map from string to any, required) — Agent-defined interrupt payload. Shape varies by agent, but typically carries `agent` (the sub-agent that raised the interrupt), `label` (human-readable prompt), and `artifacts` (proposed changes awaiting review). The orchestrator agent (`cip_orchestrator`) wraps sub-agent interrupts under a `proposals` array with `type: combined_proposal`. - `agent` (string, optional, nullable) — Agent type that raised the interrupt. Lifted from `value.agent` and surfaced at the top level for convenience; omitted when the interrupt did not declare an agent. ### AgentsStep A step in an agent workflow execution timeline. - `id` (string, required) — Unique identifier for the step. - `run_id` (string, required) — Identifier of the agent run that produced the step. - `type` (enum, required) — Category of step output. - Allowed values: `activity`, `output`, `interrupt`, `error` - `name` (string, required) — Internal step name. - `label` (string, required) — Human-readable step label. - `status` (enum, required) — Current execution status for the step. - Allowed values: `running`, `completed`, `error` - `started_at` (datetime, required) — Timestamp when the step started. - `agent` (string, optional) — Agent identifier that generated this step. - `parent_step_id` (string, optional, nullable) — Parent step identifier for nested steps. - `completed_at` (datetime, optional, nullable) — Timestamp when the step completed. - `steps` (list of AgentsStep, optional) — Nested child steps. - `result` (AgentsStepResult, optional) — Structured result payload for the step. - `artifacts` (list of AgentsArtifact, optional) — Artifacts produced by the step. ### AgentsArtifact A structured artifact produced by an agent step. - `title` (string, required) — Human-readable title for the artifact. - `type` (string, required) — Artifact type identifier. - `confidence` (string, optional, nullable) — Optional confidence label for artifact quality. - `summary` (string, optional, nullable) — Optional summary of the artifact contents. - `sources` (list of AgentsSource, optional) — Source references used to produce the artifact. ### AgentsStepResult Structured result payload for the step. - `resolution` (enum, optional) — How the interrupt was resolved. Present only on completed interrupt steps when an interrupt policy auto-approved or skipped the interrupt. Absent for manual resolutions. - Allowed values: `auto_approve`, `skip` ### AgentsSource A source reference used by an agent artifact. - `url` (string, required) — Canonical source URL. - `source_name` (string, optional) — Human-readable source name. - `source_tier` (string, optional) — Source quality or trust tier. - `source_type` (string, optional) — Source category. ## Examples **Response** ```json { "object": "list", "has_more": true, "data": [ { "object": "run", "id": "id", "created_at": "2024-01-15T09:30:00Z", "updated_at": "2024-01-15T09:30:00Z", "status": "running", "thread_id": "thread_id", "business_id": "business_id", "agent": "agent", "requester": { "id": "id", "name": "name" }, "started_at": "2024-01-15T09:30:00Z", "ended_at": "2024-01-15T09:30:00Z", "output": { "artifacts": [ { "title": "title", "type": "type" } ] }, "interrupts": [ { "id": "id", "value": { "key": "value" } } ], "steps": [ { "id": "id", "run_id": "run_id", "type": "activity", "name": "name", "label": "label", "status": "running", "started_at": "2024-01-15T09:30:00Z" } ] } ], "total_count": 1 } ``` **SDK Code** ```python import requests url = "https://api.middesk.com/v1/runs" headers = {"Authorization": "Bearer "} response = requests.get(url, headers=headers) print(response.json()) ``` ```javascript const url = 'https://api.middesk.com/v1/runs'; const options = {method: 'GET', headers: {Authorization: 'Bearer '}}; try { const response = await fetch(url, options); const data = await response.json(); console.log(data); } catch (error) { console.error(error); } ``` ```go package main import ( "fmt" "net/http" "io" ) func main() { url := "https://api.middesk.com/v1/runs" req, _ := http.NewRequest("GET", url, nil) req.Header.Add("Authorization", "Bearer ") res, _ := http.DefaultClient.Do(req) defer res.Body.Close() body, _ := io.ReadAll(res.Body) fmt.Println(res) fmt.Println(string(body)) } ``` ```ruby require 'uri' require 'net/http' url = URI("https://api.middesk.com/v1/runs") http = Net::HTTP.new(url.host, url.port) http.use_ssl = true request = Net::HTTP::Get.new(url) request["Authorization"] = 'Bearer ' response = http.request(request) puts response.read_body ``` ```java import com.mashape.unirest.http.HttpResponse; import com.mashape.unirest.http.Unirest; HttpResponse response = Unirest.get("https://api.middesk.com/v1/runs") .header("Authorization", "Bearer ") .asString(); ``` ```php request('GET', 'https://api.middesk.com/v1/runs', [ 'headers' => [ 'Authorization' => 'Bearer ', ], ]); echo $response->getBody(); ``` ```csharp using RestSharp; var client = new RestClient("https://api.middesk.com/v1/runs"); var request = new RestRequest(Method.GET); request.AddHeader("Authorization", "Bearer "); IRestResponse response = client.Execute(request); ``` ```swift import Foundation let headers = ["Authorization": "Bearer "] let request = NSMutableURLRequest(url: NSURL(string: "https://api.middesk.com/v1/runs")! as URL, cachePolicy: .useProtocolCachePolicy, timeoutInterval: 10.0) request.httpMethod = "GET" request.allHTTPHeaderFields = headers let session = URLSession.shared let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in if (error != nil) { print(error as Any) } else { let httpResponse = response as? HTTPURLResponse print(httpResponse) } }) dataTask.resume() ```