OpenRouter Usage

Fetch real-time OpenRouter usage totals and historical per-model spend. Use when the user asks for usage, spend, cost breakdown, or OpenRouter stats. Not for system health or non-LLM metrics.

This skill appears to do what it says: it uses OPENROUTER_API_KEY (and optionally OPENROUTER_MGMT_KEY) to call only openrouter.ai endpoints and print usage and model breakdowns. Before installing: (1) confirm you trust the skill owner/source (homepage is absent and owner ID looks unfamiliar); (2) prefer providing keys via environment variables rather than creating a credentials.env file in a shared filesystem; (3) be aware the registry metadata does not list the required env vars or primary credential — ask the publisher to correct that mismatch; (4) you can review scripts/stats.py yourself (it's small and readable) or run it in an isolated environment if you want to limit risk.

Type: OpenClaw Skill Name: openrouter-usage Version: 1.1.3 The skill is designed to fetch OpenRouter usage data, which it does by making authenticated API calls to `openrouter.ai`. It loads API keys from environment variables (recommended) or a `credentials.env` file as a fallback. While using a `credentials.env` file is a less secure practice for handling secrets compared to environment variables, it is explicitly documented and not indicative of malicious intent. The `SKILL.md` instructions are clear and directly relate to the skill's stated purpose, without any prompt injection attempts to subvert the agent. The Python script performs its function transparently, without unauthorized file access, network communication to external domains, or system modifications.