Skillv1.0.2

ClawScan security

WastePickupReminder · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignFeb 17, 2026, 12:48 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's files, runtime instructions, and requested resources are consistent with its stated purpose (generating formatted reminder outputs for an assistant to send); it does not request secrets or perform network I/O itself.
Guidance: This skill appears to do what it says: it generates formatted reminder outputs for your assistant to send and stores its configuration under /data/.openclaw/workspace/data/waste-reminder. Before installing, review and consider: (1) the config will contain phone numbers, chat IDs, and possibly webhook URLs — treat that folder as sensitive and limit who/what can read it; (2) the skill does not send messages itself — your AI assistant (or another integration) will do the sending, so confirm your assistant's autonomy settings if you don't want automated outbound messages; (3) you'll need to add the cron job the README suggests (cron will call the provided script every 15 minutes); and (4) inspect the included files (waste_cron.py and waste_reminder.py) locally if you want to verify behavior — there are no hidden network endpoints in the code itself. If you want stricter control, run the cron job under a restricted account and verify directory permissions for /data/.openclaw/workspace/data/waste-reminder.

Review Dimensions

Purpose & Capability: okThe skill name/description (waste collection reminders) matches the included CLI and cron scripts and the SKILL.md. No unrelated credentials, binaries, or external services are required by the code. The config/schedule file layout and channel targets are coherent with the described functionality.
Instruction Scope: okSKILL.md and the code keep scope to reading/writing config and schedule JSON and emitting formatted outputs. The scripts do not perform network calls or read unrelated system files. Note: the skill relies on the AI assistant (or an external component) to actually send messages to channels — the code only prints SEND_TO/CHANNEL/message blocks for the assistant to act on.
Install Mechanism: okNo install spec or downloads are provided; this is an instruction-and-code-only skill. No external install URLs or package installs are requested, which reduces installation risk.
Credentials: okThe skill requests no environment variables or credentials. It stores user-provided contact identifiers (phone numbers, chat IDs, and potential webhook URLs) in its workspace data directory, which is proportionate to sending reminders but is sensitive data the user should be aware of.
Persistence & Privilege: okThe skill is not set to always:true and does not try to change other skills or system configs. It asks that a cron job be scheduled to run its cron script — that is expected for periodic reminders and limited in scope to its own data directory.