audit

name	audit
description	Generate usage report for MiniMax and token optimization Use when: (1) /audit is invoked, (2) task relates to audit functionality.
user-invocable	true
context	fork

/audit

Generate usage and cost optimization report.

v2.88 Key Changes (MODEL-AGNOSTIC)

• Model-agnostic: Uses model configured in ~/.claude/settings.json or CLI/env vars
• No flags required: Works with the configured default model
• Flexible: Works with GLM-5, Claude, Minimax, or any configured model
• Settings-driven: Model selection via ANTHROPIC_DEFAULT_*_MODEL env vars

What it shows

• Total queries by model (MiniMax M2.1, lightning, Claude)
• Estimated cost savings from MiniMax usage
• Usage trends (daily/weekly)
• Optimization recommendations

Execution

# View stats
mmc --stats

# Detailed audit report
ralph audit

Cost Calculation

• Claude Sonnet: $3.00/$15.00 per 1M tokens (input/output)
• MiniMax M2.1: $0.30/$1.20 per 1M tokens (~92% savings)
• MiniMax-lightning: $0.15/$0.60 per 1M tokens (~96% savings)

Example Output

=== MiniMax Usage Audit ===
Period: Last 7 days

Model Distribution:
  MiniMax M2.1:     45 queries (60%)
  MiniMax-lightning: 20 queries (27%)
  Claude Sonnet:    10 queries (13%)

Estimated Savings:
  If all queries used Claude: ~$X.XX
  Actual cost with MiniMax:   ~$X.XX
  Savings:                    ~XX%