alphaear-search
Pass
Audited by Gen Agent Trust Hub on May 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses established search engines (DuckDuckGo, Baidu) and the Jina Search/Reader APIs for information retrieval. These interactions are standard for the skill's purpose and do not involve unauthorized data exfiltration.
- [SAFE]: Sensitive configuration, such as API keys for LLM providers (OpenAI, DeepSeek, Zai, etc.) and the Jina service, are managed strictly via environment variables, following secure coding practices.
- [SAFE]: Local data operations are restricted to a SQLite database used for search result caching and RAG functionality, with no access to sensitive system files or credentials.
- [SAFE]: External model loading for sentiment analysis and embeddings is performed through the standard Transformers library from trusted HuggingFace repositories.
- [SAFE]: The skill's architecture uses modular scripts for LLM routing and capability testing, ensuring that tool calls and reasoning tasks are directed to appropriate models securely.
Audit Metadata