data.runner.group-persist ​
Create a new runner group row (UUID-only surface)
Insert a RunnerGroup row from UUID-keyed input.
Designed as a single-state terminal workflow so the engine takes the _run_non_persistent path: only action_done is invoked, no auto_advance chain runs, and the engine never seeds state_data from input_data (which would otherwise leak the input fields into the declared output_schema — see engine.py:_run_non_persistent vs the persistent path that defaults new_state_data = dict(initial_data)).
Overview ​
| Property | Value |
|---|---|
| Workflow type | Atomic |
| Library | App-runners |
| Version | 1.0 |
Input Schema ​
| Field | Type | Required | Default | Description |
|---|---|---|---|---|
organization_uuid | uuid | Yes | — | UUID of the caller's organization, injected server-side from the authenticated request; scopes the operation to the correct tenant. |
name | string | Yes | — | Human-readable runner group name, unique per organization (and among non-deleted groups). Used to label the group and derive cloud resource names. |
backend_type | string | Yes | — | Compute backend that runs this group's runners. One of: devkit, fargate, ec2_auto_scaling, ec2_vm, gce, cloud_run, kubernetes, eks, azure_container_apps_job, azure_vm, azure_vmss, do_app_job, do_droplet, mgc_vm. Determines which cloud connection and config keys are required — call data.runner.list-provider-config-specs for the config keys per backend. |
purpose | string | Yes | — | What the runners are for. One of: github_actions, gitlab_runner, generic, agent. Use 'agent' to host Orkestia agent sessions; 'generic' for plain container jobs with no CI integration. |
integration_type | string | Yes | — | CI system this group registers runners with. One of: github, gitlab, none. 'github' requires github_installation_uuid; 'gitlab' requires gitlab_connection_uuid (+ project/group target); 'none' registers no external CI runner. |
runner_scope | string | Yes | — | Scope the runners are registered at. One of: organization, repository. 'repository' (GitHub only) requires github_repository_uuid. |
cloud_connection_uuid | uuid | No | — | UUID of the CloudConnection backing provider-backed runner groups. Required for cloud backends; omit for backend_type=devkit. |
network_profile_uuid | uuid | No | — | UUID of the NetworkProfile bound to this runner group, picked from the organization's network profiles. Optional — omit when absent. |
registry_account_uuid | uuid | No | — | UUID of the container registry account this runner group resolves images against, picked from the organization's registry accounts. Optional — falls back to default/unauthenticated image resolution when absent. |
image_pull_cloud_connection_uuid | uuid | No | — | UUID of the CloudConnection used to authenticate image pulls for this runner group, picked from the organization's connections. Optional — falls back to cloud_connection_uuid when absent. |
github_installation_uuid | uuid | No | — | UUID of the GitHub App installation bound to this runner group's GitHub integration. No list/query workflow exists in the platform today (data.github.installation.list has not been built), so this value must be supplied directly by the caller; omit when absent. |
github_repository_uuid | uuid | No | — | UUID of the GitHub repository this runner group is scoped to, picked from repositories under the organization. Optional — omit when absent (only set for repository-scoped groups). |
gitlab_connection_uuid | uuid | No | — | UUID of the GitLab CloudConnection bound to this runner group, picked from the organization's GitLab connections. Optional — omit when absent (only set for GitLab-integrated groups). |
gitlab_project_uuid | uuid | No | — | UUID of the GitLab project this runner group is scoped to. No list/query workflow exists in the platform today (data.gitlab.project.list has not been built), so this value must be supplied directly by the caller; omit when absent. |
gitlab_target_kind | string | No | — | — |
gitlab_group_path | string | No | — | — |
region | string | No | — | Cloud region the runners are provisioned in (e.g. 'us-east-1' for AWS). Required for all cloud backends (aws/gcp/azure/digitalocean/mgc); omit only for bring-your-own kubernetes. |
desired_count | integer | No | 1 | — |
min_count | integer | No | 0 | — |
max_count | integer | No | 10 | — |
description | string | No | — | — |
runner_labels | list | No | — | — |
config | dict | No | — | Backend-specific settings (e.g. ecs_cluster_name for Fargate, machine_type for GCE, container_image). Required keys vary by backend_type — call data.runner.list-provider-config-specs to get the exact keys before creating. |
coding_runtime_enabled | boolean | No | — | Enable the software-development-v1 runtime profile for this agent runner group. Requires a customer storage connection; runner.group-creation provisions and binds the private artifact bucket through the Storage workflow. |
coding_artifact_storage_connection_uuid | uuid | No | — | Customer CloudConnection used by the Storage workflow to provision private coding-artifact storage. |
coding_artifact_retention_days | integer | No | — | Recovery retention for noncurrent artifact versions, between 7 and 90 days. |
coding_artifact_capacity_bytes | integer | No | — | Required hard live-byte ceiling for the customer-owned coding-artifact bucket. Valid range: 512 MiB to 1 TiB. |
coding_artifact_storage_usd_per_gb_month | string | No | — | Customer storage price in USD per decimal GB-month, passed only to the Storage-owned provisioning workflow. |
coding_artifact_storage_pricing_revision | string | No | — | Immutable revision identifier for the supplied storage price. |
auth_type | string | No | — | — |
actor_uuid | uuid | No | — | UUID of the user or service actor who triggered this action, injected server-side from the authenticated request; used for audit/attribution only. Optional — omit for system or service-triggered actions with no human actor. |
Output Schema ​
| Field | Type | Required | Default | Description |
|---|---|---|---|---|
runner_group_uuid | uuid | No | — | UUID of the newly persisted RunnerGroup row; chain into runner.group-creation / runner.environment-provision. Also resolvable via the organization's list of runner groups. |
created_at | string | No | — | — |
error | string | No | — | — |
error_type | string | No | — | — |
States ​
| State | Initial | Terminal | Success | Auto-advance | Description |
|---|---|---|---|---|---|
done | Yes | Yes | Yes | — | Group persisted |
API Usage ​
bash
POST /api/workflows/start
Content-Type: application/json
{
"workflow_type": "data.runner.group-persist",
"initial_data": {
"organization_uuid": "value",
"name": "value",
"backend_type": "value",
"purpose": "value"
}
}