openrouter-usage
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill retrieves the OpenRouter API key from the local OpenClaw authentication store at
~/.openclaw/agents/*/agent/auth.json. This key is used to authenticate requests to the service provider.\n- [EXTERNAL_DOWNLOADS]: The script makes network requests to the official OpenRouter API (openrouter.ai) to fetch credit and usage data. This communication is restricted to the legitimate service being tracked.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted data from session log files.\n - Ingestion points:
scripts/openrouter_usage.pyreads JSONL session logs from the~/.openclawdirectory.\n - Boundary markers: No explicit delimiters or markers are used to wrap the log data in the generated report.\n
- Capability inventory: The skill has read access to local authentication and session files and performs network requests to the OpenRouter service.\n
- Sanitization: The script does not sanitize model names or token data extracted from the logs before presenting them to the agent in the usage report.
Audit Metadata