privatelm-cross-platform-llm-client
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides standard implementation patterns for integrating various AI providers (OpenAI, Anthropic, Gemini) and local inference using llama.cpp. All network operations are directed towards official API endpoints (api.openai.com, api.anthropic.com, generativelanguage.googleapis.com) or vendor-managed resources. Secret management instructions correctly advise using persistent local storage (Hive) or environment variables, avoiding hardcoded credentials. Model downloads are handled via standard HTTP requests with progress monitoring and local file system validation. No malicious patterns such as prompt injection, persistence, or data exfiltration were detected.
Audit Metadata