jobs

List jobs (paged; ?label=)

List your jobs, newest first (org admins see the whole org's jobs; members see their own). label filters to one sweep. summary: true returns the rollup instead of rows: {total, by_state, failures_by_reason, oldest_queued_age_seconds, results} - results carries each terminal job's exit code and the JSON it wrote to $GMN_RESULT_PATH, so "which of the 100 is best" is this one call. A large queued count and a growing oldest_queued_age_seconds are not a reason to cancel or to stop submitting: a sustained backlog is what raises the fleet's procurement target for the shape, so the depth you are reading is what starts the machines that drain it (each queued job's queue.demand_note says it in full).

get/preview/jobs

Request

  • Base URL: https://autoresearch.sfcompute.com
  • URL: https://autoresearch.sfcompute.com/preview/jobs
  • Auth: HTTP bearer

Query parameters

cursorstring nullable

Resume after this job id (the previous page's next_cursor) - keyset pagination, newest first.

limitinteger nullable

Max jobs to return (default 50, cap 200).

labelstring nullable

Filter to one sweep's jobs by label (an indexed filter, GMAN-24).

workspacestring

Act in this workspace (its slug). Omitted: the token's bound workspace, then the organization's default. Never ambient - there is no switch-workspace mutation on this door.

Request body

summaryboolean nullable

true = return the rollup instead of rows: {total, by_state, failures_by_reason, oldest_queued_age_seconds, results} - "which of the 100 is best" in one call.

Response

The capability's JSON result.

next_cursorstring

Pass as ?cursor= for the next page. Absent on the last page. Opaque: never parse it.

object'list' required

Changes

No recorded changes to this operation across all 1 revision of this API.