phi-cleaner

在与外部工具或合作者共享之前，使用phi-cleaner对临床文本进行去标识化处理，用于草稿、笔记及手稿摘录。适用于以下场景：(1) 从自由文本笔记中移除患者姓名、日期及标识符；(2) 生成可用于LLM提示的已消毒摘录；(3) 运行PHI检测流程以审核敏感文本；(4) 准备去标识化的附录或示例，用于手稿与演示文稿。

SKILL.md

--- frontmatter

name: phi-cleaner
description: "De-identify clinical text for drafts, notes, and manuscript excerpts using phi-cleaner before sharing with external tools or collaborators. Use when: (1) Removing patient names/dates/identifiers from free-text notes, (2) Producing sanitized excerpts for LLM prompts, (3) Running a PHI detection pass to audit sensitive text, (4) Preparing de-identified appendices or examples for manuscripts and presentations."
metadata: {"nanobot":{"requires":{"bins":["phi-clean"]},"install":[{"id":"brew","kind":"brew","formula":"drpedapati/tools/phi-cleaner","bins":["phi-clean"],"label":"Install phi-cleaner (brew)"}]}}

phi-cleaner

phi-cleaner is an optional companion CLI for text de-identification.

Use it before sending potentially sensitive clinical text into downstream tooling.

Install

bash

brew tap drpedapati/tools
brew install drpedapati/tools/phi-cleaner

Verify:

bash

phi-clean --version

Common commands

bash

# Clean direct text input
phi-clean "Patient John Smith was seen on 03/15/2024."

# Detect-only mode
phi-clean --detect "Dr. Chen at Mayo Clinic"

# File input/output
phi-clean -f note.txt -o note.cleaned.txt

# Show available models
phi-clean --models

Workflow guidance

•Run phi-clean before sharing clinical notes in chat workflows.
•Keep original and de-identified files separate.
•Validate cleaned output for context loss before publication.

Caveat

This tool helps reduce accidental PHI leakage, but it is not a legal/compliance determination on its own.