Runs & history
Reading a run's output and evidence, failure reasons, and run notifications.
Every execution of a playbook — scheduled or a test run — is recorded. Runs land in two places: the Runs card on the playbook's page (the last 50, newest first), and your home-page feed, where each run card also shows what it cost in credits and how many API calls it made.
Reading a run
Each run row shows when it ran, whether it was a Test or Scheduled run, and its status — Running, Succeeded, or Failed. Expand a row for:
- the run's output — what the playbook produced;
- one evidence block per API call: method, URL, HTTP status, and duration, expandable to the parameters sent and the response body.
Failed runs open expanded automatically, with the failure reason shown in red.
Failure reasons
Failures carry a short machine-readable reason:
| Reason | What it means |
|---|---|
creds_missing:<domain> | A mounted skill has no credential for that API domain. Add it on the skill's page and run again. |
empty_output | The run finished without producing any output — reported as a failure rather than a silent success. |
timeout | The run exceeded its time budget (five minutes). |
step_budget | The run exceeded its budget of 25 API calls without finishing. |
no_skills | The playbook had no skills mounted when the run fired. |
manifest_error:<domain> | A mounted skill's manifest couldn't be loaded. |
internal_error | Something went wrong on our side — the run was retried up to three times first. |
Automatic pause on repeated failures
Three consecutive failed scheduled runs disable the playbook so it doesn't burn credits on a broken setup — you're notified, and a success at any point resets the counter. Re-enable it from the playbook page once the cause is fixed.
Notifications
The bell in the header collects a notification for every finished scheduled run:
- Your playbook "{name}" pulse is ready. — a successful run.
- Your playbook "{name}" run failed.
- Your playbook "{name}" was disabled after repeated failures.
Each links straight to the playbook. Test runs never notify. RuneSpace sends no email about runs — the only email you'll ever get about usage is the one out-of-credits notice (see Trial & Pro).