Get the FREE Ultimate OpenClaw Setup Guide →
r

OpenRouter Usage

Verified

@rosseyre

npx machina-cli add skill @rosseyre/openrouter-usage --openclaw
Files (1)
SKILL.md
2.4 KB

OpenRouter Usage Monitor

What this skill does

Retrieves OpenRouter usage and cost data via:

  • Live totals (Today / Week / Month) from /auth/key
  • Historical per-model breakdown from /activity (completed UTC days only)

How to run (recommended)

Set environment variables (recommended) or create a credentials.env file:

export OPENROUTER_API_KEY=your_key_here
export OPENROUTER_MGMT_KEY=your_mgmt_key_here  # optional, enables model breakdown

Then execute: python3 scripts/stats.py

Alternatively, create credentials.env in the skill directory:

OPENROUTER_API_KEY=your_key_here
OPENROUTER_MGMT_KEY=your_mgmt_key_here

Fallback method (no Python)

If Python is unavailable, query endpoints directly:

Live totals curl -sS -H "Authorization: Bearer $OPENROUTER_API_KEY" https://openrouter.ai/api/v1/auth/key

Per-model activity (7d) curl -sS -H "Authorization: Bearer $OPENROUTER_MGMT_KEY" https://openrouter.ai/api/v1/activity

Configuration

Required:

  • OPENROUTER_API_KEY - Required for real-time usage totals and balance

Optional:

  • OPENROUTER_MGMT_KEY - Enables per-model spend breakdown from activity endpoint

Credentials can be provided via:

  1. Environment variables (recommended for security)
  2. credentials.env file in skill directory (fallback)

Output format

💰 OpenRouter Usage Today: $X.XX* | Week: $X.XX | Month: $X.XX Balance: $X.XX / $X.XX

Recent Models (7d): • model-name: $X.XX (N) ... * indicates live totals that may not yet appear in model breakdowns.

Edge cases

  • /activity only returns completed UTC days.
  • Today’s spend may appear in totals but not per-model data until next UTC rollover.
  • Invalid keys → 401/403.
  • Rate limiting → 429.
  • Network failures should be retried or surfaced.

Source

git clone https://clawhub.ai/rosseyre/openrouter-usageView on GitHub

Overview

OpenRouter Usage Monitor fetches real-time usage totals (Today/Week/Month) from OpenRouter /auth/key and a historical per-model spend breakdown from /activity. This lets you quantify current costs, monitor usage trends, and identify spend hotspots by model. Use this skill for usage, spend, cost breakdown, or OpenRouter stats, not for system health or non-LLM metrics.

How This Skill Works

Configure OPENROUTER_API_KEY (required) and optionally OPENROUTER_MGMT_KEY for per-model breakdown. The recommended path runs python3 scripts/stats.py to pull data from /auth/key and /activity. If Python is unavailable, you can fall back to curl calls to the same endpoints using your keys to obtain live totals and per-model activity.

When to Use It

  • User asks for today’s OpenRouter usage totals.
  • User asks for weekly or monthly OpenRouter usage totals.
  • User requests a per-model spend breakdown (requires OPENROUTER_MGMT_KEY).
  • User wants a cost overview or general OpenRouter statistics.
  • User needs a quick balance snapshot and recent model spend.

Quick Start

  1. Step 1: Set OPENROUTER_API_KEY (and optionally OPENROUTER_MGMT_KEY) in your environment or credentials.env.
  2. Step 2: Run python3 scripts/stats.py (or use curl fallback to fetch /auth/key and /activity).
  3. Step 3: Read the output under '💰 OpenRouter Usage' for Today/Week/Month and per-model spend.

Best Practices

  • Store API keys securely (prefer environment variables over credentials files).
  • Only provide OPENROUTER_MGMT_KEY when per-model breakdown is required.
  • Remember that /activity shows completed UTC days; today’s spend may lag until the next UTC rollover.
  • Anticipate rate limits (429) and implement retries with backoff.
  • Validate 401/403 responses and rotate keys if credentials are compromised.

Example Use Cases

  • Show me today’s usage totals, plus weekly and monthly totals.
  • Provide per-model spend for the last 7 days.
  • List the top spend by model for the current month.
  • Display the current balance and how much is left on the plan.
  • Explain any discrepancy between live totals and 7-day activity data.

Frequently Asked Questions

Add this skill to your agents
Sponsor this space

Reach thousands of developers