Skip to content

The open-source app everyone uses to manage agents at work

5.1 KB
Raw Blame

navigation: {"section":"Automation and structured work","sectionOrder":60,"order":10}

Routines, schedules, and triggers

Operators define repeatable work, configure its inputs and triggers, run it on demand or automatically, and inspect the execution history and resulting tasks.

Feature ID: routines.

Implementation: routine list, routine detail, trigger editor.

Agent access: companion.

Sub-features

  • definition: routines/definition — save the request, agent, project, variables, and execution context.
  • triggers: routines/triggers — configure supported schedules/webhooks or other offered triggers.
  • run-now: routines/run-now — supply inputs and create an observable routine execution.
  • history: routines/history — inspect resulting tasks and past outcomes.
  • management: routines/management — organize, pause/disable, archive, and distinguish plugin-managed routines.

How to get to it (user POV)

routine-editor

Open /routines, create one, and inspect /routines/:routineId and its edit sections.

trigger-history

Use the routine's trigger controls and Runs/Recent Runs/Audit links; lists also offer Run now. History records configuration versions, not executions.

Driving it with Playwright and the Paperclip CLI

Preconditions: follow the baseline. Use a disposable routine, deterministic test agent, and harmless target. Record schedule timezone and enabled state.

routine-editor

Automated: routine detail helper tests project-selector options, not the rendered editor. Routine service covers server contracts. Saving and running through the editor remain manual checks.

Manual: Save an agent, request, project, and a required variable. Reload and inspect the saved definition. Run now with a distinguishable input, follow the resulting task, and verify its output and history entry. Check a missing required input.

Variables are detected from {{name}} placeholders; mark a detected variable Required and verify missing input prevents Run routine. Configuration can be checked without model credentials. A successful subprocess run alone does not complete the generated task or establish a useful task output.

trigger-history

Automated: trigger wizard is component coverage; routine integration is server integration rather than a provider-wide browser proof.

Manual: Set a near-term disposable schedule or authorized test webhook, verify one eligible execution and its source trigger, then disable it and confirm further invocations stop. Inspect a failed execution without losing its history. Remove the temporary trigger after testing.

Gotchas

  • Saving a schedule does not prove the scheduler invoked it.
  • Timezone, overlap/concurrency rules, and plugin ownership affect behavior.
  • A routine run’s task must still respect pause, access, and budget gates.
  • Managed idle hosts may use the routine wake protocol to sleep between supported future runs. The host must save a durable alarm and verify the same generation before stopping. Local execution and catch-up remain unchanged. Unsupported triggers and other pending work keep it awake.

Latest skill references

Routine instructions retain stable skill mentions. After a project source syncs an upstream change, the next execution uses the latest installed version without waiting on GitHub. Record the resolved version before/after and verify a resumed execution keeps its original version. Private skill references require current project access for both the agent and its responsible person.

Storybook verification

run-now

Capability ID: routines/run-now.

Scope: ui Storybook: Mount RoutineRunVariablesDialog through actual RoutineDetail with required/default inputs, project/workspace overrides, pending, blocked and failed submission. RunButton-only no-op examples do not supply/run inputs. Production: POST /routines/{id}/run must validate variables, create one attributed execution/task and obey pause/access/budget gates; inspect actual task outcome separately. Sources: Source: RoutineRunVariablesDialog.tsx, RoutineDetail.tsx, routines.ts. Tests: RoutineRunVariablesDialog.test.tsx, routines-service.test.ts, routines-e2e.test.ts

Documentation scope: Routines, cases, and status cards

All three capabilities have recipes. Trigger/provider/timezone combinations, custom case schemas and summarizer quality need deeper cases.