Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Calculates Claude Max subscription usage from OpenClaw session data using a reverse-engineered credits formula. Shows weekly budget consumption, 5-hour rate limit window status, and per-session token breakdowns across Opus, Sonnet, and Haiku models.
Exposes the actual credits math behind Claude Max plans so you can see real consumption before hitting an unexpected rate limit.
When to use it
- Checking how much weekly budget remains before hitting limits
- Identifying which sessions consumed the most credits
- Monitoring the 5-hour sliding window before rate limiting
- Comparing credit usage across Pro, 5x, and 20x plans
- Getting a per-model token breakdown for a specific session
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: how much of my weekly Claude budget have I used?
- 1Run claude-usage.py with saved config to fetch session JSONL data
- 2Calculate credits consumed per session using the input/output token rates for each model
- 3Sum credits across all sessions in the current weekly window
- 4Compare total against the plan budget and render a progress bar
- 5Check the 5-hour sliding window and flag if approaching the rate limit
Weekly summary showing 12,400,000 / 41,666,700 credits used (29%), 5-hour window at 18%, and top sessions ranked by consumption
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.