Homemade Machine Learning Skill

Machine learning skill: find, explain, and implement ML algorithms with interactive Jupyter Notebook links. Covers linear regression, logistic regression, ne...

Security Scan

The skill is internally consistent with its description: it is an instruction-only ML education helper that prints explanations, code snippets, and links to GitHub/nbviewer; it does not request credentials or perform unexpected network/file operations.

Purpose & Capability

Name/description promise (explain algorithms, provide notebooks and code) matches the provided SKILL.md and the included shell script, which resolves queries and prints explanations, snippets, and links to trekhleb's GitHub and nbviewer.

Instruction Scope

Runtime instructions direct the agent/user to run the local script with specific subcommands (explain, notebook, code, path, list). The script only reads its CLI args and prints content/URLs; it does not attempt to read unrelated files, environment variables, or transmit data to unexpected endpoints.

Install Mechanism

No install spec is provided (instruction-only plus a shipped script). Nothing is downloaded or written to disk by an installer; the included script is plain shell and not obfuscated.

Credentials

No environment variables, credentials, or config paths are required. The skill does not request tokens or secrets and only references public GitHub/nbviewer URLs.

Persistence & Privilege

always is false, agent invocation is normal default, and the skill does not modify other skills or system config. It has no elevated persistence or system-wide privileges.

Assessment

This skill appears to be a benign educational helper that prints algorithm explanations, Python snippets, and links to public GitHub/nbviewer notebooks. Before installing, review the included script (scripts/ml-notebook-finder.sh) yourself — it is plain shell and only prints text and URLs. When you follow the notebook links, remember they point to external content (GitHub / nbviewer); review any external notebooks before running code from them. Do not supply secrets or credentials to this skill (it does not need any). If you want to be extra cautious, run the script in a restricted environment or inspect its full output before using linked notebooks.

Machine learning skill: learn, explain, and implement ML algorithms from scratch. Based on trekhleb/homemade-machine-learning (MIT, 22k+ ⭐)

📦 Install: clawhub install homemade-machine-learning-skill

5 algorithms · 11 interactive notebooks · math explained · Python code included

Commands

explain — 解释算法原理 + 数学 + 代码

bash scripts/ml-notebook-finder.sh explain "linear regression"
bash scripts/ml-notebook-finder.sh explain "neural network"
bash scripts/ml-notebook-finder.sh explain "kmeans"

notebook — 获取交互式 Jupyter Notebook 链接

bash scripts/ml-notebook-finder.sh notebook "logistic regression"
bash scripts/ml-notebook-finder.sh notebook "anomaly detection"

code — 获取 Python 实现代码片段

bash scripts/ml-notebook-finder.sh code "linear regression"
bash scripts/ml-notebook-finder.sh code "kmeans"

path — 生成学习路径（按难度排序）

bash scripts/ml-notebook-finder.sh path beginner
bash scripts/ml-notebook-finder.sh path intermediate
bash scripts/ml-notebook-finder.sh path advanced

list — 列出所有算法

bash scripts/ml-notebook-finder.sh list

Algorithms

Algorithm	Type	Notebooks	Use Case
linear regression	supervised	3	price prediction, forecasting
logistic regression	supervised	4	classification, MNIST
neural network (MLP)	supervised	2	image recognition, deep learning
k-means	unsupervised	1	clustering, segmentation
anomaly detection	unsupervised	1	fraud detection, monitoring

Source

MIT License — Original author: trekhleb Indexed by BytesAgain — AI skill discovery platform