openclaw-cron-maintenance

name	openclaw-cron-maintenance
description	Use when maintaining OpenClaw cron-managed bridge jobs such as Telegram Digest or AgentMail Inbox, especially when updating schedules, fixing sync-openclaw-cron-jobs.sh, validating jobs.json, or recovering from a hanging `openclaw cron list` / cron CLI path.

OpenClaw Cron Maintenance

Use this skill for server-side cron maintenance around bridge-style jobs (telethon-digest, agentmail-email, similar).

Core rule

On this deployment, do not trust openclaw cron list/add/remove as the primary path. The gateway CLI can hang while the scheduler and cron store still work.

Prefer:

1. inspect the cron store JSON directly
2. patch only the managed jobs for one prefix
3. back up the store before changes
4. restart openclaw-openclaw-gateway-1
5. validate via jobs.json + container health

Default cron-store locations

Check in this order:

• /opt/openclaw/config/cron/jobs.json
• /home/deploy/.openclaw/cron/jobs.json

Allow override with OPENCLAW_CRON_STORE.

Safe workflow

1. Read the managed prefix and intended schedules from the repo sync script.
2. Read the current cron store and extract only jobs matching that prefix.
3. Back up the store to jobs.json.bak-<unix_ts>.
4. Replace only the managed jobs for that prefix.
5. Preserve other jobs untouched.
6. Restart openclaw-openclaw-gateway-1.
7. Wait until the gateway returns to healthy.
8. Validate the store contains the expected names, schedule.expr, and enabled=true.

Useful checks

Check gateway/container health:

ssh -i ~/.ssh/id_rsa "$OPENCLAW_HOST" \
  'docker ps --format "table {{.Names}}\t{{.Status}}" | grep -E "openclaw-openclaw-gateway-1|telethon-digest-cron-bridge|agentmail-email-bridge"'

Inspect cron store:

ssh -i ~/.ssh/id_rsa "$OPENCLAW_HOST" \
  'sudo cat /opt/openclaw/config/cron/jobs.json 2>/dev/null || sudo cat /home/deploy/.openclaw/cron/jobs.json'

When editing sync scripts

• Keep them idempotent.
• Patch only one managed prefix per script.
• Preserve existing id / createdAtMs when the same job name still exists.
• Recompute nextRunAtMs.
• Restart the gateway after patching the store.
• Never rewrite unrelated cron jobs.

Validation target

After a successful change, confirm:

• gateway is healthy
• bridge container is Up
• managed jobs exist in jobs.json
• expected cron expressions are present
• no unrelated jobs were removed