file-cleanup
Session-based file cleanup for privacy compliance. Use to delete temporary files, uploaded clinical notes, generated outputs, and expired logs to maintain HIPAA-friendly operations.
浏览 RooseveltAdvisors 维护的已发布 AI 技能,比较用途与说明,并通过来源仓库了解安装方式。
已收录技能
9收录技能的仓库 Stars 合计
0仓库 Forks
0平均每个技能的仓库 Stars
0同一仓库中的多个技能可能共享 Stars,合计包含重复仓库数据。
Session-based file cleanup for privacy compliance. Use to delete temporary files, uploaded clinical notes, generated outputs, and expired logs to maintain HIPAA-friendly operations.
Pydantic schema validation for agent outputs. Use to validate JSON against TableOfContents, ClinicalSummary, TreatmentPlan, and EvaluatorFeedback schemas before accepting agent responses.
JSON-based structured logging for audit trails and debugging. Use for logging all agent operations, quality metrics, errors, and execution times with daily rotation and automatic cleanup.
Detect clinical note sections (HPI, ROS, Assessment, Plan) using regex patterns and LLM topic segmentation. Generates ToC with accurate byte offsets.
Phi-4 LLM interaction skill for generating text completions via Ollama API. Use for all LLM inference tasks including section detection, summarization, recommendation generation, and quality evaluation.
Hybrid search (embedding + BM25) for retrieving relevant clinical note passages. Use for finding source evidence to support claims in summaries and recommendations.
Contextual Retrieval implementation for RAG - chunks clinical notes with LLM-generated context prepended to each chunk before embedding. Improves citation accuracy by 49% per Anthropic research.
Extract and validate citations with Jaccard overlap calculation. Ensures all claims have verifiable source evidence with ≥0.7 alignment threshold.
ChromaDB vector database client for storing and retrieving text embeddings with hybrid search (dense + sparse). Use for RAG operations, contextual retrieval, and similarity search in clinical notes.
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库,同一仓库的多个技能可能共享相同数值;它反映仓库关注度,不代表单个技能的安装量或质量评分。