Token Usage Monitor
PassAudited by ClawScan on May 1, 2026.
Overview
The skill appears aligned with its stated purpose of locally tracking token usage, with the main things to notice being persistent local usage history and optional periodic monitoring integration.
This looks reasonable for a token usage monitor. Before installing, be aware that it stores usage history locally and that the optional startup/heartbeat integration would make monitoring recurring rather than purely manual.
Findings (2)
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.
Your token usage history may remain on disk across sessions until you delete it.
The skill discloses persistent local storage of usage history. This is expected for a token monitor, but users should know it leaves a record of sessions, models, timestamps, and token counts.
The script automatically creates and manages a data file at `~/.openclaw/token_usage.json` to store usage data.
Install only if you are comfortable with local usage history being stored, and delete or protect the data file if the usage metadata is sensitive.
If you add the heartbeat integration, the agent may periodically check usage thresholds and generate reports without a new manual command each time.
The guide recommends optional persistent/periodic monitoring via heartbeat logic. This fits the monitoring purpose and is user-directed, but it can make the agent perform recurring checks if the user enables it.
Update your HEARTBEAT.md to include periodic token usage checks: ... - Check token usage thresholds hourly ... - Generate daily usage report at 9:00 AM
Only add the startup or heartbeat integration if you want ongoing monitoring; keep it scoped to token metrics and remove the heartbeat entries if you no longer need them.