lightrag
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a guide for implementing the LightRAG library, a tool for graph-based knowledge extraction and querying.
- [EXTERNAL_DOWNLOADS]: The skill references the
lightrag-hkupackage from the official PyPI registry and the source code from the HKU Data Science Lab's GitHub repository (HKUDS/LightRAG). These are established open-source and academic resources. - [CREDENTIALS_UNSAFE]: The documentation correctly advises users to store sensitive API keys (OpenAI, Anthropic, Gemini) and database credentials in a
.envfile rather than hardcoding them. The examples provided use safe placeholders. - [INDIRECT_PROMPT_INJECTION]: As a RAG-based tool, the skill is designed to ingest and process external documents. While this introduces a surface for indirect prompt injection common to all RAG systems, the skill does not exhibit specific vulnerabilities or unsafe processing logic.
Audit Metadata