Requirements and responsibilities
Confirm that the deployed build exposes OpenClaw automations and the operator has application admin permission. The upstream scheduler depends on a running Gateway; xCloud does not document a dedicated automation panel.
OpenClaw automations · OpenClaw automation CLI · Managed OpenClaw Hosting · How to Deploy OpenClaw with xCloud
Choose public sources, one recipient, a time zone and a harmless first run.
OpenClaw automations · OpenClaw automation CLI · Managed OpenClaw Hosting · How to Deploy OpenClaw with xCloud
Illustrative situation
Illustrative scenario: an owner wants a weekday digest of two public status pages sent to the approved Telegram account.
Choose the approach
Use the OpenClaw application scheduler only after finding it in the deployed interface or getting a documented operator handoff. Keep xCloud server cron separate from application-owned automation.
OpenClaw automations · OpenClaw automation CLI · Managed OpenClaw Hosting · How to Deploy OpenClaw with xCloud
Dashboard and application procedure
Follow these steps yourself, or use the scoped AI handoff below for supported hosting operations.
Step 1 of 5
Define the digest contract
- Where
- Owner runbook
- Permissions
- Authorized OpenClaw owner or application administrator
- Inputs
- Two public URLs, weekday time, time zone and Telegram recipient
- Action
- Specify a two-bullet digest with direct source links, one message to the named Telegram recipient per occurrence, and no other writes.
- Expected result
- A bounded job specification exists.
- Verify
- Have the recipient confirm source list and local delivery time.
- If it fails
- If no owner can identify a missed or duplicate run, do not enable recurrence.
Sources: Managed OpenClaw Hosting · OpenClaw automations
Step 2 of 5
Verify app scheduler access
- Where
- Deployed OpenClaw application
- Permissions
- Authorized OpenClaw owner or application administrator
- Inputs
- Current automation controls and admin identity
- Action
- Have an authorized app administrator verify the Gateway and `openclaw automations` availability through the documented xCloud SSH handoff if granted; list existing jobs before changes. Do not assume an xCloud cron UI creates app jobs.
- Expected result
- The feature and permission path are known.
- Verify
- Locate existing jobs before creating a new one.
- If it fails
- If no scheduler surface is available on this hosted build, stop and leave the digest manual.
Sources: How to Deploy OpenClaw with xCloud · OpenClaw automations · OpenClaw automation CLI
Step 3 of 5
Create one named job
- Where
- OpenClaw automation control
- Permissions
- Authorized OpenClaw owner or application administrator
- Inputs
- Unique job name, schedule and destination
- Action
- Have the app administrator use the current `openclaw automations` create syntax for a recurring isolated job, with the agreed time zone and approved Telegram destination. Permit only the named Telegram delivery; keep source reads public and all other writes out of scope.
- Expected result
- One job is listed with the expected schedule and recipient.
- Verify
- Compare name, time zone, payload and destination with the runbook.
- If it fails
- If a similar job already exists, update or retire it instead of creating a duplicate.
Sources: How to Deploy OpenClaw with xCloud · OpenClaw automations · OpenClaw automation CLI
Step 4 of 5
Inspect one due run
- Where
- OpenClaw job history and Telegram
- Permissions
- Authorized OpenClaw owner or application administrator
- Inputs
- Scheduled occurrence ID or time window
- Action
- Wait for a due run or use an authorized test run, then inspect `openclaw automations runs <job-id>` and compare it with the delivered Telegram message and links.
- Expected result
- Exactly one complete digest is delivered for the occurrence.
- Verify
- Check run status, recipient, timestamp, two bullets and source links.
- If it fails
- If delivery fails or duplicates, pause the job and reconcile history before retrying.
Sources: How to Deploy OpenClaw with xCloud · OpenClaw automations · OpenClaw automation CLI
Step 5 of 5
Assign failure ownership
- Where
- Runbook and application job list
- Permissions
- Authorized OpenClaw owner or application administrator
- Inputs
- Owner, missed-run threshold and pause path
- Action
- Document who checks job history and how to pause or resume after a provider or channel outage.
- Expected result
- The job has a named operator and recovery rule.
- Verify
- Have the operator locate the job and explain how to detect a duplicate.
- If it fails
- If history or pause control is unavailable, treat the schedule as unverified and disable it.
Sources: How to Deploy OpenClaw with xCloud · OpenClaw automations
Maintenance
At each due occurrence, compare one job-history entry with exactly one Telegram message, inspect source links, and flag a missed or duplicate delivery.
Recovery decisions
If a run fails or duplicates, pause the named automation, preserve its job ID and run history, and reconcile which occurrence was delivered before a retry. Resume only after one test run passes.
AI handoff
Use the named xCloud dashboard for hosted controls and verify application tools in the deployed OpenClaw version before use. An xCloud hosting MCP write operation for these actions is not documented.
Supported scope
- Configure an OpenClaw application behavior app · manual
Upstream OpenClaw documentation describes application features. xCloud does not document every app configuration surface or pre-enabled tool. Verify the feature and permissions in the deployed build before editing.
Checkpoint: Use a harmless positive and negative test; preserve current configuration and stop if the feature is unavailable.
How to Deploy OpenClaw with xCloud · OpenClaw security and access control
Copyable agent brief
Manual checkpoints
- An authorized application admin confirms the deployed Gateway and current `openclaw automations` CLI before creating the job.
- The recipient checks one message against `openclaw automations runs <job-id>` and pauses duplicates.
Feature coverage
- task setup and decision (covered): Defines prerequisites and an explicit hosted/application handoff. Define the digest contract Verify app scheduler access
- task acceptance and failure (covered): One scheduled occurrence produces exactly one linked two-bullet digest to the approved recipient, with a matching run-history entry. Failure: Unavailable scheduler, incorrect time zone, missing history, failed delivery or duplicate delivery blocks recurring use. Create one named job Inspect one due run Assign failure ownership