Diagnose, audit, and repair OpenClaw hosts when an agent, gateway, plugin, cron, model route, memory engine, channel, approval policy, or update looks broken...
Documents
ground-control
Try itVerify OpenClaw after upgrades against an operator-maintained model, cron, and channel ground truth. Use for report-only config integrity, provider liveness, cron integrity, session smoke tests, and channel checks; require explicit approval before applying repairs.
What it does
Verify OpenClaw after upgrades against an operator-maintained model, cron, and channel ground truth. Use for report-only config integrity, provider liveness, cron integrity, session smoke tests, and channel checks; require explicit approval before applying repairs.
The skill document
ground-control
Post-upgrade verification for OpenClaw. Keeps your system honest after every upgrade.
🛠️ Installation
1. Ask OpenClaw (Recommended)
Tell OpenClaw: "Install the ground-control skill." Installation does not authorize runtime config or cron changes.
2. Manual Installation (CLI)
If you prefer the terminal, run:
openclaw skills install @jonathanjing/ground-control
Permissions & Privileges
This skill requires the following OpenClaw capabilities:
gateway config.get— read current config (all phases)gateway config.patch— apply an operator-approved config repair (Phase 1 only)cron list/cron update— verify cron jobs and apply operator-approved repairs (Phase 3)sessions_spawn— smoke test sessions (Phase 2, 4, 5)message send— channel liveness test + summary report (Phase 5)
Change-control behavior: Every run is report-only by default. Present a redacted repair plan and obtain explicit operator approval immediately before any gateway config.patch or cron update.
Security & Redaction: This skill enforces a Zero-Secret Logging protocol.
- Immediate Redaction: Sensitive nodes (
auth,plugins) are stripped from memory after fetching runtime config. - Redacted Drift: Mismatches in sensitive fields are reported as
[REDACTED_SENSITIVE_MISMATCH]. - Functional Validation: API keys are tested through functional calls (Phase 2), never through literal comparison.
- No Persistence: Literal credentials are never written to
memory/files or messaging channels.
Environment variables: None.
When to use
- After running
openclaw updateornpm install -g openclaw@latest - When you suspect config drift (model changed, cron broken, channel down)
- Periodic health check via
/verifycommand
Setup
- Copy
{baseDir}/templates/MODEL_GROUND_TRUTH.mdto your workspace root - Fill in your actual config values (models, cron jobs, channels)
- (Optional) For large configs, split into
refs/ground-truth/*.mdsub-files - Add the GROUND_TRUTH sync rule to your AGENTS.md (see README)
- Run
/verifyto test
Files
{baseDir}/templates/MODEL_GROUND_TRUTH.md— Ground truth template (copy to workspace root){baseDir}/scripts/post-upgrade-verify.md— Agent execution prompt for 5-phase verification{baseDir}/scripts/UPGRADE_SOP.md— Upgrade standard operating procedure
Related skills
Operate and extend the read-only OpenClaw dashboard backend, frontend tabs, and security model.
Use when upgrading OpenClaw or recovering from partial updates, gateway failures, migrations, service drift, channel failures, or version mismatch.
Use the OpenClaw cron tool for scheduling reminders, delayed follow-ups, and recurring periodic checks. Covers one-shot and recurring schedules, session targeting, delivery modes, and wake events.
Route OpenClaw CLI requests to the right command family and profile, then verify the outcome.
Manage OpenClaw cron jobs — list, pause, resume, maintenance mode, and health checks.