Get all runs by agent
List runs for a specific workflow.
Supports filtering by status, search_key, and error_code. All filters are combined with AND logic.
Unindexed free-text searches can paginate only through the first 1,000 matching runs. Split larger exports into narrower created_at_start/created_at_end windows or use a complete run/task ID.
search_key
Case-insensitive substring search across: workflow run ID, parameter key, parameter description, run parameter value, extra HTTP headers and webhook callback URL. A run is returned if any of these fields match. Soft-deleted parameter definitions are excluded from key/description matching. A complete browser profile ID, browser session ID or credential ID matches the run that used it exactly (no substring match). A complete workflow run ID matches that run. A complete task ID matches the workflow run the task ran in, when this list includes that run, or the standalone task itself where standalone tasks are listed. A complete browser profile, browser session, workflow run or task ID is matched against those identifiers only, not the text fields. A credential ID matches when it is the run’s sequential credential, the credential the run selected from a pool or fell back to, or the credential bound by a credential parameter on the run’s workflow version when the run was created and the run recorded no selection for it. The workflow title and workflow permanent ID are matched as well.
error_code
Exact-match filter on the error_code field inside each task’s errors JSON array. A run matches if any of its tasks contains an error with a matching error_code.
include_child_runs
Child runs — runs this workflow started from inside another workflow run — are excluded by default, so the history shows only top-level runs. Pass include_child_runs=true to list them as well; every run carries a parent_workflow_run_id that is null for top-level runs.
Authorizations
Your Skyvern API key from https://app.skyvern.com/settings.
Path Parameters
Query Parameters
Page number for pagination.
1 <= x <= 1000000Number of runs to return per page.
1 <= x <= 1000Filter by one or more run statuses.
created, queued, running, failed, terminated, canceled, timed_out, completed, paused Case-insensitive substring search across: workflow run ID, parameter key, parameter description, run parameter value, extra HTTP headers and webhook callback URL. A run is returned if any of these fields match. Soft-deleted parameter definitions are excluded from key/description matching. A complete browser profile ID, browser session ID or credential ID matches the run that used it exactly (no substring match). A complete workflow run ID matches that run. A complete task ID matches the workflow run the task ran in, when this list includes that run, or the standalone task itself where standalone tasks are listed. A complete browser profile, browser session, workflow run or task ID is matched against those identifiers only, not the text fields. A credential ID matches when it is the run's sequential credential, the credential the run selected from a pool or fell back to, or the credential bound by a credential parameter on the run's workflow version when the run was created and the run recorded no selection for it. The workflow title and workflow permanent ID are matched as well.
500"login_url"
Exact-match filter on the error_code field inside each task's errors JSON array. A run matches if any of its tasks contains an error with a matching error_code. Error codes are user-defined strings set during workflow execution.
500"INVALID_CREDENTIALS"
Only include runs created at or after this UTC timestamp (ISO 8601).
Only include runs created strictly before this UTC timestamp (ISO 8601).
Filter by run tags. Each term is a label (production), a group (env:*), or a group:label (env:prod). Repeat the param or comma-separate (?tags=env:prod,env:staging). AND across distinct terms, OR within a group's labels (?tags=customer:acme,env:prod,env:staging -> customer=acme AND env in (prod, staging)). A label term matches the value across any/no group. Matches current tag values only.
20"env:prod"
Include child workflow runs — runs started from inside another workflow run. Excluded by default so the history shows only top-level runs.
Response
Successful Response
created, queued, running, failed, terminated, canceled, timed_out, completed, paused Which layer of the seed-precedence chain seeded a run's browser (provenance).
Resolved at run setup, before any browser creation, for all run types (C-semantics). A serialized Save & Reuse run re-resolves once more after the sequential wait, so a profile an earlier run in its lane saved while it was queued becomes own_memory rather than a stale pick.
- override: explicit request browser_profile_id (one-run-only pick via API)
- picked: the workflow's explicit profile pick (workflows.browser_profile_id) — "always start here"
- own_memory: the workflow's own auto-profile (no pick + persist_browser_session)
- credential: the run's selected credential's profile (rotation-aware; also the empty-own boot)
- fresh: no seed profile
- degraded_fresh: a resolved profile failed to load; ran fresh
override, picked, own_memory, credential, fresh, degraded_fresh One-based number of the current workflow run attempt
Whether another attempt is scheduled for this workflow run
Timestamp when the next workflow run attempt is scheduled
Attempts recorded for this workflow run
RESIDENTIAL, US-CA, US-NY, US-TX, US-FL, US-WA, RESIDENTIAL_ES, RESIDENTIAL_IE, RESIDENTIAL_GB, RESIDENTIAL_IN, RESIDENTIAL_JP, RESIDENTIAL_FR, RESIDENTIAL_DE, RESIDENTIAL_NZ, RESIDENTIAL_ZA, RESIDENTIAL_AR, RESIDENTIAL_AU, RESIDENTIAL_BR, RESIDENTIAL_TR, RESIDENTIAL_CA, RESIDENTIAL_MX, RESIDENTIAL_IT, RESIDENTIAL_NL, RESIDENTIAL_PH, RESIDENTIAL_KR, RESIDENTIAL_SA, RESIDENTIAL_ISP, NONE Browser settings copied from the workflow version when the run was created
How a workflow run was initiated.
- manual: User clicked "Run" in the UI
- mcp: First-party MCP client or Skyvern CLI request
- api: Direct API call to the run endpoint
- scheduled: Triggered by a cron schedule
- webhook: Triggered by an external system via the webhook endpoint
- event: Triggered by an external event source
- job_recipe_extract: Launched by a job recipe extract request
- job_recipe_apply: Launched by a job recipe apply request
- seo_recipe_submit: Launched by an SEO recipe directory submission
manual, mcp, api, scheduled, webhook, event, job_recipe_extract, job_recipe_apply, seo_recipe_submit ID of the user who started the run

