Med-LLM-Wiki

管理医疗领域患者临床咨询知识库，结构化存储患者信息、问答、干预方案及循证概念，支持精准细节检索与推理。

This skill is coherent with its stated goal of creating and querying a structured medical wiki, but it explicitly instructs the agent to ingest and persist patient-identifiable clinical data without any privacy or security guidance. Before installing or enabling it, you should: 1) ensure the agent runs in a secure, access-controlled environment (HIPAA-compliant if applicable); 2) define exactly where the wiki/ files will be stored and who can access them (avoid writing to shared or cloud-exposed paths); 3) add or require mechanisms for de-identification or explicit patient consent if you will store PHI; 4) consider adding encryption-at-rest, access logging, and retention/deletion policies; 5) restrict autonomous invocation or require user confirmation before creating or returning any patient-specific content; and 6) if you need stricter control, request the author add explicit config options for secure storage and auditability. If you cannot guarantee secure handling of sensitive health data in your environment, do not enable this skill.

Type: OpenClaw Skill Name: med-llm-wiki Version: 1.0.0 The skill bundle provides a structured framework for an AI agent to manage a medical knowledge base (Wiki) focused on patient consultations and clinical reasoning. The instructions in SKILL.md define a clear four-layer data model (patients, Q&A, interventions, and concepts) and provide templates for organizing medical data into Markdown files with frontmatter. There is no evidence of malicious intent, data exfiltration, or unauthorized system access; the logic is entirely focused on document processing and information retrieval within a local 'wiki/' directory.