openclaw-audit-watchdog
PassAudited by ClawScan on May 1, 2026.
Overview
This appears to be a transparent audit scheduler, but it creates a recurring job and sends security reports to configured DM/email recipients that users should review.
This skill looks coherent for automated OpenClaw security monitoring. Before installing or invoking it, verify the release source, confirm the DM/email recipients and SMTP settings, review the cron schedule and persisted environment variables, and avoid enabling suppressions or optional git pull unless you intentionally want those behaviors.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The audit job may keep running daily until the user disables or removes it.
The recurring unattended job is explicitly disclosed and matches the audit-watchdog purpose, but it means the skill will continue operating after setup.
when invoked it creates or updates an unattended `openclaw cron` job
Before enabling, confirm the schedule, recipients, and host, and make sure you know how to list and remove the OpenClaw cron job if needed.
Security findings about your agents or host could be sent to the wrong DM handle, channel, email address, or SMTP relay if misconfigured.
The skill sends audit output outside the local process through DM and optional email. This is disclosed and purpose-aligned, but security audit reports can contain sensitive findings.
External delivery: reports go to the configured DM target and optionally to the configured email recipient, so review those recipients before enabling automation
Use trusted recipients only, double-check DM and email destinations, and prefer a trusted local or secured mail relay for email delivery.
Installing from a release archive requires trusting that release and its integrity.
The documented standalone installation downloads and extracts a release archive. The docs also advise verifying source and integrity, so this is a review note rather than a hidden supply-chain concern.
curl -sSL "https://github.com/prompt-security/clawsec/releases/download/$VERSION_TAG/openclaw-audit-watchdog.skill" -o watchdog.skill unzip watchdog.skill
Install from the intended publisher, verify the release source and checksum/signature where available, and avoid enabling optional automatic git pulls unless you trust the repository.
If suppressions are enabled carelessly, important warnings could be demoted in reports.
The skill supports persistent suppression/allowlist configuration that can change how future audit findings are presented. The artifacts show explicit opt-in gates and visible suppressed findings.
Suppression is **opt-in with defense in depth**: the audit pipeline requires BOTH a CLI flag AND a config-file sentinel before any finding is suppressed.
Only enable suppressions with a reviewed configuration file, keep clear reasons and dates, and periodically audit the suppression list.