Openrouter Usage Analytics

Overview

This skill demonstrates building usage analytics for OpenRouter to track costs, token usage, model popularity, and performance metrics over time.

Prerequisites

• OpenRouter integration with logging
• Data storage for metrics (database, time-series DB, or analytics service)

Instructions

1. Instrument API calls: Add middleware that captures timestamp, model, prompt/completion token counts, cost, latency, and status for every request
2. Store metrics efficiently: Write metrics to a time-series database (InfluxDB, TimescaleDB) or append-only table with proper indexing on timestamp and model
3. Build aggregation queries: Create queries for daily/weekly cost by model, average latency trends, token usage distribution, and error rates
4. Create a dashboard: Visualize key metrics — total spend over time, cost per model, average latency, requests per hour, and error rate trends
5. Set up automated reports: Schedule daily or weekly summary emails/alerts with top-line metrics and any anomalies detected

Output

• Analytics pipeline capturing cost, latency, and usage per request
• Dashboard with charts for spend, token usage, model distribution, and errors
• Automated weekly report summarizing key usage trends and anomalies

Error Handling

See ${CLAUDE_SKILL_DIR}/references/errors.md for full error reference.

Examples

See ${CLAUDE_SKILL_DIR}/references/examples.md for runnable code samples.

Resources

• OpenRouter Documentation
• OpenRouter Models
• OpenRouter API Reference
• OpenRouter Status