Reminder Guardian
Analysis
The skill mostly matches its reminder purpose, but its code appears to call a time-helper path outside the packaged skill, so it should be reviewed before installation.
Findings (3)
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.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
SKILL_ROOT = Path(__file__).resolve().parents[2] TIME_HELPER = SKILL_ROOT / "scripts" / "time_helper.py" ... result = subprocess.run(cmd, capture_output=True, text=True)
The packaged helper is supplied as scripts/time_helper.py, but reminder_guard.py is itself under scripts/. Using parents[2] points above the skill root before appending scripts/time_helper.py, so the normal helper call can target an unprovided external path rather than the bundled helper.
Create the cron job: Copy the printed blueprint and run `openclaw cron add`
The skill asks the user to create a scheduled job from a generated blueprint. This is purpose-aligned and manual, but scheduled jobs are persistent actions that should be reviewed.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
LOG_PATH = REPO_ROOT / "memory" / "reminder-log.json" ... "message": args.message, "note": args.note,
Reminder messages and optional notes are stored in a persistent local JSON file and later reused for listings and cron blueprints.
