# Show what a session's turn is doing or got done

Source: <https://everruns.com/docs/api/operations/get_session_work/>

[Docs](https://everruns.com/docs/)/[API reference](https://everruns.com/docs/api/)/[Sessions & workspaces](https://everruns.com/docs/api/groups/sessions-workspaces/)/Show what a session's turn is doing or got done

# Show what a session's turn is doing or got done

Preview

Manage execution sessions and rooms, and their files, tasks, storage, databases and managed sandboxes.

Public APIEverruns Cloud and self-hosted Platform

GET`/v1/sessions/{session_id}/work`

See [authentication](https://everruns.com/docs/sre/runbooks/authentication/) for credentials and permissions.

## Parameters

### Path Parameters

**session_id**

required

string

Prefixed public identifier

## Responses

200Success

### 200

Success

Media typeapplication/json

Response of `GET /v1/sessions/{session_id}/work`.

object

**report**

One of:

* object

  The report, or `null` when the session has not run a turn.

  object

  **agent_name**

  Agent name snapshot at turn start.

  string | null

  **attention**

  One of:

  * object

    The open request when the work waits on a person.

    object

    **kind**

    required

    What kind of answer is needed.

    string

    Allowed values: approval question action

    **prompt**

    required

    One line naming what is asked (“Run git push”).

    string

    **tool_call_ids**

    required

    Open client-side call ids that carry the request.

    Array\<string>

  * null

    null

  **duration_ms**

  Elapsed time: to the end when finished, to the read time while running.

  integer | null format: int64

  **ended_at**

  When the work ended.

  string | null format: date-time

  **headline**

  required

  One line, always present: the current step while working, the request while waiting, the result when finished.

  string

  **model**

  Model of the latest generation.

  string | null

  **omitted_steps**

  Steps run but no longer listed.

  integer format: int32

  **plan**

  One of:

  * object

    The plan the agent shared, if any.

    object

    **current**

    The item in progress, in its “doing” form when the tool has one.

    string | null

    **done**

    required

    Items done or skipped.

    integer format: int32

    **items**

    required

    Plan items in order.

    Array\<object>

    One checklist line in \[`BackgroundProgress::steps`].

    object

    **status**

    required

    State of one checklist line.

    string

    Allowed values: pending in_progress done skipped

    **title**

    required

    string

    **total**

    required

    Items in total.

    integer format: int32

  * null

    null

  **result**

  One of:

  * object

    What finished work produced. Absent while working.

    object

    **artifacts**

    Files, links, pull requests and sessions the work produced.

    Array\<object>

    Typed link to something the task produced.

    object

    **name**

    required

    string

    **path**

    Session VFS path, when the artifact lives in the session filesystem.

    string | null

    **type**

    required

    Artifact type: “file”, “url”, “session”, “pr”, etc.

    string

    **url**

    External URL, when the artifact lives elsewhere.

    string | null

    **checks**

    How the agent checked the work (“ran the 12 billing tests”), when it reported.

    string | null

    **facts**

    Measured facts, in display order.

    Array\<object>

    One measured fact.

    object

    **kind**

    required

    Which fact.

    string

    Allowed values: files_changed issues tasks_started steps

    **value**

    required

    Its value.

    integer format: int64

    **outcome**

    One of:

    * string

      The agent’s own verdict, when it reported.

      string

      Allowed values: done failed

    * null

      null

    **summary**

    One or two sentences on what changed.

    string | null

    **summary_source**

    One of:

    * string

      Where the summary came from.

      string

      Allowed values: agent reply

    * null

      null

  * null

    The agent’s own verdict, when it reported.

    string

    Allowed values: done failed

  * string

    null

  * null

    Where the summary came from.

    string

    Allowed values: agent reply

  * string

    null

  * null

    null

  **started_at**

  When the work started.

  string | null format: date-time

  **state**

  required

  Where the work stands.

  string

  Allowed values: working needs_you done failed stopped

  **steps**

  required

  Steps in order, newest last, bounded by \[`MAX_REPORT_STEPS`].

  Array\<object>

  One step of the work: a tool call, labelled by its narration.

  object

  **duration_ms**

  How long the step took, once finished.

  integer | null format: int64

  **id**

  required

  The tool call id.

  string

  **label**

  required

  What the step is doing (“Running tests”) or what came out of it (“Read AGENTS.md”).

  string

  **progress**

  One of:

  * object

    Live progress while running.

    object

    **current**

    Items done so far, when the tool knows a count.

    integer | null format: int64

    **label**

    The tool’s latest status line.

    string | null

    **total**

    Items in total, when the tool knows a count.

    integer | null format: int64

  * null

    null

  **started_at**

  When the step started.

  string | null format: date-time

  **state**

  required

  Where the step stands.

  string

  Allowed values: running done issue

  **tool_name**

  required

  Tool name.

  string

  **turn_id**

  required

  The turn the report describes.

  string

* null

  The open request when the work waits on a person.

  object

  **kind**

  required

  What kind of answer is needed.

  string

  Allowed values: approval question action

  **prompt**

  required

  One line naming what is asked (“Run git push”).

  string

  **tool_call_ids**

  required

  Open client-side call ids that carry the request.

  Array\<string>

* object

  null

* null

  The plan the agent shared, if any.

  object

  **current**

  The item in progress, in its “doing” form when the tool has one.

  string | null

  **done**

  required

  Items done or skipped.

  integer format: int32

  **items**

  required

  Plan items in order.

  Array\<object>

  One checklist line in \[`BackgroundProgress::steps`].

  object

  **status**

  required

  State of one checklist line.

  string

  Allowed values: pending in_progress done skipped

  **title**

  required

  string

  **total**

  required

  Items in total.

  integer format: int32

* object

  null

* null

  What finished work produced. Absent while working.

  object

  **artifacts**

  Files, links, pull requests and sessions the work produced.

  Array\<object>

  Typed link to something the task produced.

  object

  **name**

  required

  string

  **path**

  Session VFS path, when the artifact lives in the session filesystem.

  string | null

  **type**

  required

  Artifact type: “file”, “url”, “session”, “pr”, etc.

  string

  **url**

  External URL, when the artifact lives elsewhere.

  string | null

  **checks**

  How the agent checked the work (“ran the 12 billing tests”), when it reported.

  string | null

  **facts**

  Measured facts, in display order.

  Array\<object>

  One measured fact.

  object

  **kind**

  required

  Which fact.

  string

  Allowed values: files_changed issues tasks_started steps

  **value**

  required

  Its value.

  integer format: int64

  **outcome**

  One of:

  * string

    The agent’s own verdict, when it reported.

    string

    Allowed values: done failed

  * null

    null

  **summary**

  One or two sentences on what changed.

  string | null

  **summary_source**

  One of:

  * string

    Where the summary came from.

    string

    Allowed values: agent reply

  * null

    null

* object

  The agent’s own verdict, when it reported.

  string

  Allowed values: done failed

* null

  null

* string

  Where the summary came from.

  string

  Allowed values: agent reply

* null

  null

* string

  null

* null

  Live progress while running.

  object

  **current**

  Items done so far, when the tool knows a count.

  integer | null format: int64

  **label**

  The tool’s latest status line.

  string | null

  **total**

  Items in total, when the tool knows a count.

  integer | null format: int64

* object

  null

* null

  null

### Example

```json
{
  "report": {
    "agent_name": "Billing fixer",
    "attention": {
      "kind": "approval",
      "prompt": "Deploy to staging"
    },
    "duration_ms": 133000,
    "ended_at": "2026-10-23T12:02:13Z",
    "headline": "Step 3 of 4 · Running `cargo test`",
    "model": "gpt-6.2-sol",
    "omitted_steps": 0,
    "plan": {
      "current": "Run the tests",
      "done": 2,
      "items": [
        {
          "status": "pending"
        }
      ],
      "total": 4
    },
    "result": {
      "checks": "Ran the 12 billing tests",
      "facts": [
        {
          "kind": "files_changed",
          "value": 3
        }
      ],
      "outcome": "done",
      "summary": "Fixed the proration rounding and added a test.",
      "summary_source": "agent"
    },
    "started_at": "2026-10-23T12:00:00Z",
    "state": "working",
    "steps": [
      {
        "duration_ms": 4200,
        "id": "call_01",
        "label": "Ran `cargo test`, 12 tests passed",
        "progress": {
          "current": 7,
          "label": "Rendering image 7 of 12",
          "total": 12
        },
        "started_at": "2026-10-23T12:00:00Z",
        "state": "running",
        "tool_name": "bash"
      }
    ],
    "turn_id": "turn_01933b5a00007000800000000000001"
  }
}
```

404Session or turn not found

### 404

Session or turn not found

FetchcURL

Request

```js
const url = 'https://app.everruns.com/api/v1/sessions/example/work';
const options = {method: 'GET'};


try {
  const response = await fetch(url, options);
  const data = await response.json();
  console.log(data);
} catch (error) {
  console.error(error);
}
```

Request

```sh
curl --request GET \
  --url https://app.everruns.com/api/v1/sessions/example/work
```

Additional language examples require JavaScript. The default request example is available above.

200404

200 · Success

```json
{
  "report": {
    "agent_name": "Billing fixer",
    "duration_ms": 133000,
    "ended_at": "2026-10-23T12:02:13Z",
    "headline": "Step 3 of 4 · Running `cargo test`",
    "model": "gpt-6.2-sol",
    "omitted_steps": 0,
    "started_at": "2026-10-23T12:00:00Z",
    "state": "working",
    "steps": [],
    "turn_id": "turn_01933b5a00007000800000000000001"
  }
}
```

404 · Session or turn not found

No response body example is declared for this status.

## Matching endpoints