What URLs does GitHub Extract accept?

It accepts repo/blob and repo/tree URLs from public GitHub repositories and converts them to raw URLs for content retrieval.

How do I save the downloaded content?

Use the --save option; it saves the file to a temporary path with a proper filename and returns the path.

What dependencies and OS are supported?

Requires uv and wget. Supported OS include darwin, linux, and win32.

GitHub Extract

@guoqiao

npx machina-cli add skill @guoqiao/gh-extract --openclaw

Files (1)

SKILL.md

1.0 KB

GitHub Extract

Extract content from a GitHub url.

Use this skill when the user types /gh-extract or asks to extract/download/summarize a GitHub url.

What it does

Accepts an GitHub url, could be repo/tree/blob.
Convert the url to github raw url.
Extract file content from the raw url or save to a temp path.

Requirements

uv
wget

Usage

# print file content to stdout
uv run --script ${baseDir}/gh_extract.py <url>

# save file to a temp path, with a proper filename
uv run --script ${baseDir}/gh_extract.py <url> --save

Notes

only works for public repo.
url can be repo/tree/blob
for repo/tree, will try to get README.md or SKILL.md or README.txt

Source

git clone https://clawhub.ai/guoqiao/gh-extractView on GitHub

Overview

GitHub Extract fetches file contents from a GitHub URL (repo, tree, or blob) by converting it to a raw URL and downloading the content. It can either print to stdout or save to a temp path for later processing, and it works with public repositories using uv and wget.

How This Skill Works

Provide a GitHub URL to the gh_extract.py script via uv run. The tool converts the URL to a raw GitHub URL, downloads the file content, and outputs it or saves it to a temp path with a proper filename. If the URL points to a repo/tree, it will attempt to fetch README.md, SKILL.md, or README.txt as a fallback.

When to Use It

Extract a single file's raw content from a public GitHub URL (repo/blob).
Fetch content from a repo/tree URL and fall back to README.md, SKILL.md, or README.txt.
Save the extracted content to a temporary path for downstream processing.
Automate retrieval of docs or skill definitions from multiple GitHub repos without cloning.
Quickly preview a file's contents in stdout during scripting or debugging.

Quick Start

Step 1: Ensure uv and wget are installed on your system.
Step 2: Print file content to stdout: uv run --script ${baseDir}/gh_extract.py <url>
Step 3: Save to a temp path: uv run --script ${baseDir}/gh_extract.py <url> --save

Best Practices

Only use with public GitHub repositories.
Prefer explicit paths (repo/blob or repo/tree/path) when possible.
Use --save to cache the file with a meaningful, temporary filename.
Validate the downloaded content is non-empty before downstream use.
Ensure uv and wget are installed and accessible on the system.

Example Use Cases

Pull README.md from a public project to generate a quick project outline.
Extract SKILL.md from a repo to populate an AI agent skills directory.
Retrieve a config file from a repository to feed into an automation script.
Preview a file's content in stdout to verify correctness before processing.
Save a downloaded file to a temp path for subsequent parsing in a workflow.

Frequently Asked Questions

Add this skill to your agents