Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Tracks AI API usage by logging each call with model, token counts, and cost to a JSONL file. Provides CLI reports broken down by model, time period, or use case, and suggests cheaper model alternatives based on task complexity.
Requires no external service or API key, so all cost data stays local and is queryable instantly without a dashboard login.
When to use it
- Checking total AI spend for the month
- Comparing costs across Claude, GPT, and other models
- Auditing which tasks consume the most tokens
- Deciding when to swap a model for a cheaper one
- Generating weekly spend summaries for a team
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: show AI spending for the last 7 days by model
- 1Run `python report.py --days 7 --by-model`
- 2Parse the JSONL log for entries within the date range
- 3Aggregate token counts and costs per model
- 4Format results into a spend breakdown table
A per-model cost summary covering the past 7 days with token totals and dollar amounts