Run any Skill in Manus with one click

checkup-report-interpreter

Interprets physical examination reports by parsing PDF files into structured data, identifying abnormalities with severity grading, and generating clinical explanations with health recommendations. Supports year-over-year comparison of two reports. Use when the user uploads a checkup report or asks for help understanding lab results.

Run Skill in Manus

Overview

Install command

npx skills add https://github.com/vitaclaw/vitaclaw --skill checkup-report-interpreter

Copy and paste this command into Claude Code to install the skill

Source

vitaclaw/vitaclaw

Stars17

Forks3

UpdatedMarch 23, 2026 at 16:00

File Explorer

3 files

SKILL.md

readonly

name	checkup-report-interpreter
description	Interprets physical examination reports by parsing PDF files into structured data, identifying abnormalities with severity grading, and generating clinical explanations with health recommendations. Supports year-over-year comparison of two reports. Use when the user uploads a checkup report or asks for help understanding lab results.
version	1.0.0
user-invocable	true
argument-hint	[report <pdf> \| parse <pdf> \| extract <pdf> \| compare <pdf1> <pdf2>]
allowed-tools	Read, Grep, Glob, Write, Edit
metadata	{"openclaw":{"emoji":"🏥","category":"health-scenario"}}

Checkup Report Interpreter

Automatically parses physical examination report PDFs into structured data, then uses LLM for abnormality identification, clinical interpretation, and health recommendation generation.

Features

Smart PDF Parsing: Uses PyMuPDF to extract text from checkup reports, supporting multi-page reports
Structured Extraction: LLM identifies all examination items (laboratory tests, imaging findings, physical examination), returning standardized JSON
Abnormality Grading: Automatically determines abnormality severity based on reference ranges (urgent / important / moderate / minor), with 40+ built-in common indicator reference values
Clinical Interpretation: Groups abnormal indicators by organ system and explains their clinical significance in plain language
Health Recommendations: Generates personalized health recommendations and suggested follow-up items
Annual Comparison: Supports item-by-item comparison of two reports, highlighting new abnormalities, worsening trends, and improvements

Usage

# Full report interpretation
python checkup_report_interpreter.py report checkup_report.pdf

# Parse PDF text only
python checkup_report_interpreter.py parse checkup_report.pdf

# Extract structured examination items
python checkup_report_interpreter.py extract checkup_report.pdf

# Compare two years of reports
python checkup_report_interpreter.py compare 2026_checkup.pdf 2025_checkup.pdf

Environment Variables

Variable	Required	Description
`OPENROUTER_API_KEY`	Yes	OpenRouter API key
`OPENROUTER_BASE_URL`	No	API endpoint (default: `https://openrouter.ai/api/v1/chat/completions`)
`LLM_MODEL`	No	Model name (default: `google/gemini-2.5-flash`)

More from this repository

same repository

annual-checkup-advisor

vitaclaw/vitaclaw

Orchestrates comprehensive annual checkup interpretation by coordinating report parsing, lab interpretation, family history analysis, genetic risk scoring, TCM constitution assessment, and guideline lookup. Use when the user uploads a checkup report or asks for help interpreting physical examination results.

2026-03-2317

apple-health-digest

vitaclaw/vitaclaw

Parse Apple Health export (export.xml) and correlate wearable data with medical treatment history — optimized for oncology patients tracking health metrics across treatment cycles. Tracks heart rate, blood pressure, SpO2, weight, daily steps, and sleep, then overlays these onto treatment periods to reveal patterns. Generates per-metric reports and a treatment correlation analysis. Supports HealthKit MCP server integration. Works with records from medical-record-organizer. Trigger phrases: 分析我的苹果健康数据, 导入Apple Health数据, 化疗期间体重变化, 治疗期间心率, 看我的步数趋势, analyze Apple Health export, import health data, track weight during chemo, heart rate during treatment, activity trend.

2026-03-2317

blood-pressure-tracker

vitaclaw/vitaclaw

Records and classifies blood pressure readings per ACC/AHA 2017 guidelines, detects morning surge, analyzes diurnal variation, and generates monthly statistics. Use when the user logs BP readings or asks about blood pressure trends.

2026-03-2317

chronic-condition-monitor

vitaclaw/vitaclaw

Monitors multiple chronic disease indicators (BP, glucose, HbA1c, lipids, uric acid, creatinine, eGFR, liver function) against Chinese clinical guidelines. Detects abnormal trends, metabolic syndrome, and generates visit summaries. Use when the user tracks lab results or manages multiple chronic conditions.

2026-03-2317

health-memory

vitaclaw/vitaclaw

Centralized health memory hub — manages daily logs and per-item longitudinal tracking files under memory/health/. WHEN TO USE: After any health skill records data, or when answering health questions that need historical context.

2026-03-2317

health-timeline

vitaclaw/vitaclaw

Query and visualize medical history for any patient — optimized for tracking cancer treatment progression and chronic disease management. Displays chronological treatment history, lab value trends (tumor markers CEA/CA-199/AFP/CA-125, CBC, kidney/liver function) with ASCII charts, imaging findings, and treatment response (CR/PR/SD/PD) across all lines of therapy. Flags abnormal values and generates a print-ready timeline for doctor visits. Correlates with Apple Health wearable data when available. Works with records organized by medical-record-organizer. Trigger phrases: 看我的病情时间线, CEA变化趋势, 最近肿瘤标志物, 化疗后指标, 查看治疗历史, 疗效评估, show health timeline, tumor marker trend, disease progression, treatment history, lab value trends.

2026-03-2317

Source

vitaclaw

vitaclaw/vitaclaw

View GitHub Repository View Creator Repositories

Install command

Download

Run Skill in Manus

Useful forSOC

Software DevelopersComputer and Mathematical Occupations15-1252L4

name	checkup-report-interpreter
description	Interprets physical examination reports by parsing PDF files into structured data, identifying abnormalities with severity grading, and generating clinical explanations with health recommendations. Supports year-over-year comparison of two reports. Use when the user uploads a checkup report or asks for help understanding lab results.
version	1.0.0
user-invocable	true
argument-hint	[report <pdf> \| parse <pdf> \| extract <pdf> \| compare <pdf1> <pdf2>]
allowed-tools	Read, Grep, Glob, Write, Edit
metadata	{"openclaw":{"emoji":"🏥","category":"health-scenario"}}

Checkup Report Interpreter

Automatically parses physical examination report PDFs into structured data, then uses LLM for abnormality identification, clinical interpretation, and health recommendation generation.

Features

Smart PDF Parsing: Uses PyMuPDF to extract text from checkup reports, supporting multi-page reports
Structured Extraction: LLM identifies all examination items (laboratory tests, imaging findings, physical examination), returning standardized JSON
Abnormality Grading: Automatically determines abnormality severity based on reference ranges (urgent / important / moderate / minor), with 40+ built-in common indicator reference values
Clinical Interpretation: Groups abnormal indicators by organ system and explains their clinical significance in plain language
Health Recommendations: Generates personalized health recommendations and suggested follow-up items
Annual Comparison: Supports item-by-item comparison of two reports, highlighting new abnormalities, worsening trends, and improvements

Usage

# Full report interpretation
python checkup_report_interpreter.py report checkup_report.pdf

# Parse PDF text only
python checkup_report_interpreter.py parse checkup_report.pdf

# Extract structured examination items
python checkup_report_interpreter.py extract checkup_report.pdf

# Compare two years of reports
python checkup_report_interpreter.py compare 2026_checkup.pdf 2025_checkup.pdf

Environment Variables

Variable	Required	Description
`OPENROUTER_API_KEY`	Yes	OpenRouter API key
`OPENROUTER_BASE_URL`	No	API endpoint (default: `https://openrouter.ai/api/v1/chat/completions`)
`LLM_MODEL`	No	Model name (default: `google/gemini-2.5-flash`)