Skill Detail
Track coding-agent token spend and leaderboard drift across local session logs with Tokscale
Aggregate local coding-agent usage logs into spend, token, and trend views before quotas or budgets surprise the team.
Monitoring & AlertsMulti-Framework
Monitoring & Alerts
Multi-Framework
Security Reviewed
⭐ 2k GitHub stars
⬇ 64.6k/wk npm
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill track-coding-agent-token-spend-and-leaderboard-drift-with-tokscale
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Supported coding-agent local logs, Node.js or Bun for install, optional browser or TUI
Install & setup
Install the CLI with the upstream package instructions, point it at the default local agent log locations, then run Tokscale to inspect token, cost, and trend views or submit leaderboard data if desired.
Author
junhoyeo
Publisher
Individual
Last updated
Apr 20, 2026
Quick brief
Use Tokscale when you need an agent-adjacent monitoring pass over local Claude Code, Codex, OpenClaw, Cursor, Gemini, and similar session logs. It ingests per-tool histories, normalizes token and cost data, and surfaces budget drift, usage trends, and optional leaderboard reporting from one place. Invoke this instead of using each coding product’s built-in UI when the job is cross-agent cost tracking from local artifacts rather than one tool’s native analytics. The boundary is token and spend observability across coding-agent logs, not a generic AI platform or standalone dashboard listing.