Llm Eval Router

• Analyzed: May 11, 2026, 4:48 AM
• Findings: 1
• Version: 1.2.2
• Hash: 1e918bd5d08207596ae02a37d343d3d529cd26ad7cd0cec021549e0663d6a230
• Source: palm
• External report: View on VirusTotal

Findings (1)

The skill defines an LLM evaluation and routing pipeline that requires outbound network access to multiple cloud providers (Anthropic, OpenAI, Gemini) and the use of sensitive API keys. While these capabilities are aligned with the stated purpose of benchmarking local models, the requirement for background process persistence (cron/systemd) and broad network access constitutes a high-risk profile. No evidence of intentional malice, obfuscation, or prompt injection was found in SKILL.md or _meta.json, but the risky capabilities are significant.