一键在 Manus 中运行任何 Skill

habit-troubleshoot

星标6

分支2

更新时间2026年7月21日 10:36

Diagnose and fix HABIT runtime errors — covers preprocess, habitat clustering, feature extraction, and ML/comparison failures. Use when the user pastes a Python traceback, when a `habit ...` CLI command fails, when an output file is missing or corrupt, or when results look biologically wrong (degenerate clusters, all-NaN features, AUC=0.5). Triggers on phrases like "报错", "出错", "不工作", "为什么", "error", "failed", "traceback", "crash", "AUC 太低", "没有生成结果".

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

lichao312214129

lichao312214129/HABIT

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

HABIT Troubleshooting

When something goes wrong, don't guess — match the symptom against the playbook below. Each error file in references/ covers one HABIT command.

How the agent should use this skill

Identify which step failed (preprocess / habitat / extract / ml).
Open the corresponding error file:
- Preprocess errors → references/errors_preprocess.md
- Habitat errors → references/errors_habitat.md
- Extraction errors → references/errors_extraction.md
- ML / comparison errors → references/errors_ml.md
- Cross-cutting recovery → references/recovery_playbook.md
Search the file for the symptom (error message text, file content).
Apply the suggested fix. If it works, log what was wrong (helps catch recurring issues).

When to ask the user for more info

If the user just says "it failed" without details, ask for:

Which habit ... command they ran
The full traceback (last ~20 lines)
Whether the user has changed the config since the last successful run
Output of python skills/habit-quickstart/scripts/check_environment.py

Do NOT recommend solutions blindly. The error message is almost always enough to localize the cause.

Universal first response to any failure

Run the appropriate validator script — it catches 80% of issues without needing the traceback:

If failure was during...	Run this
`habit preprocess`	`validate_preprocess_output.py <out_dir>`
`habit get-habitat`	`validate_habitat_output.py <out_dir>`
`habit extract`	`inspect_feature_csv.py` on each output CSV
`habit model`	`inspect_feature_csv.py` on the input CSV
`habit compare`	`inspect_feature_csv.py` on each prediction CSV

Hand-off to other skills

If error reveals a wrong config: regenerate from the corresponding specialist skill's references/ template.
If error reveals missing data: hand off to habit-quickstart's check_data_layout.py.
If error is environment-level (ImportError, command not found): ask user to rerun check_environment.py and report back.

Known unfixable cases

These cases are NOT software bugs — explain to the user and stop:

Image registration failed because images are anatomically incompatible (e.g. trying to register brain MRI to abdominal CT). Solution: re-acquire or use a different fixed_image.
Tumor too small for habitat clustering (< 100 voxels). Solution: the user must either redraw a larger ROI or accept that habitat analysis is inappropriate here.
All clinical labels are 0 (or all 1). Solution: there is no model to train; this is a data problem.

同仓库更多 Skills

同仓库

habit-quickstart

lichao312214129/HABIT

Entry-point router for the HABIT (Habitat Analysis Biomedical Imaging Toolkit) package. Use this FIRST whenever a user mentions HABIT, habitat analysis, tumor sub-region clustering, intra-tumor heterogeneity, "生境分析", "亚区聚类", "habitat segmentation", or wants a multi-step radiomics workflow. Verifies environment, validates data layout, then hands off to the right specialist skill.

2026-07-216

habit-feature-extraction

lichao312214129/HABIT

Extract quantitative features from HABIT habitat maps — traditional radiomics, whole-habitat radiomics, per-habitat radiomics, MSI (Most Significant Intensity), and ITH (Intra-Tumor Heterogeneity) scores. Use when the user has habitat .nrrd maps and wants per-subject feature CSVs ready for ML. Triggers on "提取生境特征", "MSI 特征", "ITH 异质性", "habitat features", "extract radiomics from habitats". Runs `habit extract`.

2026-07-066

habit-habitat-analysis

lichao312214129/HABIT

Generate tumor habitat (sub-region) maps from medical images using voxel-level clustering. Use when the user wants to identify intra-tumor heterogeneity zones, perform supervoxel clustering, generate habitat .nrrd maps, run kinetic DCE habitat, or use voxel-radiomics texture clustering. Triggers on "生境分析", "亚区聚类", "habitat", "supervoxel", "kinetic", "DCE 分期生境", "intra-tumor heterogeneity". Runs `habit get-habitat`.

2026-07-066

habit-machine-learning

lichao312214129/HABIT

Train, predict, or k-fold cross-validate ML classifiers on habitat / radiomics feature CSVs using HABIT. Supports LogisticRegression, RandomForest, XGBoost, SVM, MLP, AutoGluon, and 12+ feature selection methods (LASSO, mRMR, RFECV, ICC, correlation, ANOVA, ...). Use when the user wants to build a prediction model from feature CSVs. Triggers on "训练模型", "建模", "K折交叉验证", "k-fold", "LASSO", "feature selection", "predict mode". Runs `habit model` or `habit cv`.

2026-07-066

habit-model-comparison

lichao312214129/HABIT

Compare multiple trained classification models with publication-quality plots — ROC, DCA, calibration, precision-recall, DeLong's AUC test. Use when the user has prediction CSVs from 2+ models and wants side-by-side comparison. Triggers on "模型比较", "ROC 对比", "DeLong 检验", "决策曲线", "校准曲线", "model comparison", "compare AUC", "DCA". Runs `habit compare`.

2026-07-066

habit-radiomics

lichao312214129/HABIT

Extract traditional PyRadiomics features (firstorder, shape, GLCM, GLRLM, GLSZM, NGTDM, GLDM) from medical images at the whole-tumor ROI level — no habitat segmentation. Use when the user wants classical radiomics features without doing habitat analysis. Triggers on "传统影像组学", "PyRadiomics", "radiomics features", "shape features", "GLCM". Runs `habit radiomics`.

2026-07-066

name

habit-troubleshoot

description

HABIT Troubleshooting

When something goes wrong, don't guess — match the symptom against the playbook below. Each error file in references/ covers one HABIT command.

How the agent should use this skill

Identify which step failed (preprocess / habitat / extract / ml).
Open the corresponding error file:
- Preprocess errors → references/errors_preprocess.md
- Habitat errors → references/errors_habitat.md
- Extraction errors → references/errors_extraction.md
- ML / comparison errors → references/errors_ml.md
- Cross-cutting recovery → references/recovery_playbook.md
Search the file for the symptom (error message text, file content).
Apply the suggested fix. If it works, log what was wrong (helps catch recurring issues).

When to ask the user for more info

If the user just says "it failed" without details, ask for:

Which habit ... command they ran
The full traceback (last ~20 lines)
Whether the user has changed the config since the last successful run
Output of python skills/habit-quickstart/scripts/check_environment.py

Do NOT recommend solutions blindly. The error message is almost always enough to localize the cause.

Universal first response to any failure

Run the appropriate validator script — it catches 80% of issues without needing the traceback:

If failure was during...	Run this
`habit preprocess`	`validate_preprocess_output.py <out_dir>`
`habit get-habitat`	`validate_habitat_output.py <out_dir>`
`habit extract`	`inspect_feature_csv.py` on each output CSV
`habit model`	`inspect_feature_csv.py` on the input CSV
`habit compare`	`inspect_feature_csv.py` on each prediction CSV

Hand-off to other skills

If error reveals a wrong config: regenerate from the corresponding specialist skill's references/ template.
If error reveals missing data: hand off to habit-quickstart's check_data_layout.py.
If error is environment-level (ImportError, command not found): ask user to rerun check_environment.py and report back.

Known unfixable cases

These cases are NOT software bugs — explain to the user and stop:

Image registration failed because images are anatomically incompatible (e.g. trying to register brain MRI to abdominal CT). Solution: re-acquire or use a different fixed_image.
Tumor too small for habitat clustering (< 100 voxels). Solution: the user must either redraw a larger ROI or accept that habitat analysis is inappropriate here.
All clinical labels are 0 (or all 1). Solution: there is no model to train; this is a data problem.