← 返回 Skills 市场

Beta Lead Scoring

Name: Beta Lead Scoring
Author: 1477009639zw-blip

作者 1477009639zw-blip · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

总下载

当前安装

版本数

在 OpenClaw 中安装

/install betaleadscore

功能描述

AI-powered B2B lead scoring model. Predicts conversion probability for potential customers using machine learning (LightGBM + SHAP). CSV upload or API integr...

使用说明 (SKILL.md)

Lead Scoring Model

AI-powered B2B lead scoring using LightGBM + SHAP for interpretability.

Usage

python3 score.py --input leads.csv --output scores.csv

Features

CSV upload → scored leads with conversion probability
Top features driving each score (SHAP)
Ranked priority list
Pipeline: LightGBM → SHAP → actionable insights

Input CSV Format

company_size,industry,page_views,email_opens,form_fills,job_title_score
50,tech,120,5,2,8
200,finance,45,2,0,5

Output

lead_id,score,probability,top_factor,risk_level
1,0.85,85%,page_views,hot
2,0.32,32%,low_engagement,cold

Notes

MIT-0 License | Requires: python3, lightgbm, shap, pandas

安全使用建议

This package is not malicious, but it's misleading: it advertises a LightGBM+SHAP model while the included score.py is a simple rule-based scorer. Before installing or using it: (1) review the Python script yourself — it only requires pandas/numpy and does not call external services or handle secrets; (2) if you expect a trained LightGBM model or SHAP explanations, ask the author for the model artifact and the real code that computes SHAP values; (3) avoid installing heavy ML packages unless you actually need them; and (4) run the script on non-sensitive sample data first to confirm behavior. If you need a true ML-backed lead scorer, obtain the trained model and verify how feature importances/SHAP are computed and stored.

能力评估

⚠ Purpose & Capability

The description and SKILL.md advertise LightGBM + SHAP and interpretability, but score.py contains a simple rule-based implementation that never imports or uses lightgbm or shap. Declaring heavyweight ML libraries in 'Notes' is disproportionate to the actual code and could mislead users about capabilities or required setup.

ℹ Instruction Scope

Runtime instructions are narrow and clear (run python3 score.py --input ...). They do not instruct the agent to read unrelated files, access secrets, or make network calls. However, the SKILL.md claims features (SHAP explanations, model pipeline) that are not implemented by the provided script, so the instructions and claimed scope are inconsistent.

ℹ Install Mechanism

There is no install spec (lowest risk). The SKILL.md lists dependencies (lightgbm, shap, pandas) but the package does not include installation steps and the script only uses pandas/numpy. This is not an install risk but is inconsistent and could lead users to install unnecessary/large ML packages.

✓ Credentials

The skill requests no environment variables, credentials, or config paths. The code reads only the supplied CSV and writes an output CSV — no secrets or unrelated system access are requested.

✓ Persistence & Privilege

always:false and no installation or background persistence. The skill does not modify other skills or system settings and does not request persistent presence.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install betaleadscore
安装完成后，直接呼叫该 Skill 的名称或使用 /betaleadscore 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

init

元数据

Slug betaleadscore

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题