openrouter-usage
Skill41 starsChanged 7 months agoMITView on GitHub
Query OpenRouter API costs, credits, and usage breakdown by model, provider, and date for the last 30 days. Triggers on API spending, check costs, openrouter usage, how much have I spent.
- Reads credentials
Files
OpenRouter Usage
Query OpenRouter API costs and generate usage reports via the command line.
When to Use
- Checking OpenRouter spend over the last 30 days
- Breaking down costs by model, provider, or date
- Exporting usage data as CSV or JSON for accounting
- Auditing which models and API keys consume the most credits
Prerequisites
OPENROUTER_API_KEYenvironment variable (or in a project.envfile)- The key must be a Management key — create one at https://openrouter.ai/settings/management-keys
- Regular provisioning keys return 401/403 on analytics endpoints
Usage
# Full 30-day summary with credits balance
python3 ~/.claude/skills/openrouter-usage/scripts/openrouter_usage.py
# Filter to a single date
python3 ~/.claude/skills/openrouter-usage/scripts/openrouter_usage.py --date 2026-02-01
# Export as CSV (for spreadsheets)
python3 ~/.claude/skills/openrouter-usage/scripts/openrouter_usage.py --csv
# Raw JSON output (for piping to jq)
python3 ~/.claude/skills/openrouter-usage/scripts/openrouter_usage.py --json
# Skip credits lookup (activity only)
python3 ~/.claude/skills/openrouter-usage/scripts/openrouter_usage.py --no-creditsOutput
The default report includes:
- Credits balance: total purchased, used, remaining
- Spend by model: per-model cost with token counts (prompt, completion, reasoning)
- Spend by provider: aggregated by provider (OpenAI, Anthropic, Google, etc.)
- Spend by date: daily spend with visual bar chart
API Endpoints
Both endpoints are read-only GETs authenticated via Bearer token:
| Endpoint | Purpose |
GET /api/v1/credits | Total credits purchased and used |
GET /api/v1/activity | Last 30 days of usage grouped by model/endpoint/provider |
The activity endpoint returns per-row: date, model, provider_name, usage, byok_usage_inference, requests, prompt_tokens, completion_tokens, reasoning_tokens.
In this repository99 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- agents-md-managerSkill
- architecture-md-builderSkill
- autoresearchSkill
- claude-agent-sdkSkill
- claude-md-managerSkill
- claude-usageSkill
- dag-typesafeSkill
- fableSkill
- glossarySkill
- react-best-practicesSkill
- skill-optimizerSkill
- stop-slopSkill
- tamarruSkill
- test-harness-auditorSkill
- user-model-builderSkill
- atmosphere-shaderSkill
- component-gallerySkill
- conductor-motionSkill
- design-auditSkill
- design-critiqueSkill
- design-mdSkill
- design-polishSkill
- gemini-claude-resonanceSkill
- gemini-forgeSkill
- gpt-atelierSkill
- grainientSkill
- image-forgeSkill
- image-wellSkill
- meshySkill
- mindnode-uiSkill
- minoan-frontend-designSkill
- nano-banana-proSkill
- overwatch-slidedeckSkill
- paper-designSkill
- particle-swarm-simSkill
- pellicolaSkill
- photo-to-slack-emojiSkill
- pretextSkill
- recordlySkill
- rocaille-shaderSkill
- scroll-cinemaSkill
- shadcnSkill
- shapeSkill
- smolvlmSkill
- speak-responseSkill
- sprite-forgeSkill
- threejs-particle-canvasSkill
- vellum-editorialSkill
- webgpu-threejs-tslSkill
- agent-browserSkill
- beautiful-mermaidSkill
- classical-887Skill
- claude-peersSkill
- cloudflareSkill
- codex-ctoSkill
- codex-orchestratorSkill
- disc-forgeSkill
- figma-mcpSkill
- linkedin-exportSkill
- The first 60 shown.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

