OneFileLLM is a Python CLI tool that aggregates content from local files, GitHub repos, web pages, PDFs, YouTube transcripts, and more, then outputs a structured XML and copies it to the clipboard for LLM ingestion.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
OneFileLLM is a command-line tool that automates data aggregation from various sources (local files, GitHub repos, web pages, PDFs, YouTube transcripts, etc.) and combines them into a single, structured XML output that's automatically copied to your clipboard for use with Large Language Models.
How it works
The project accepts multiple input sources (paths, URLs, or aliases) via a CLI. It supports local files, GitHub repositories and issues, web pages, academic sources (ArXiv, DOIs, PMIDs), and multimedia sources (YouTube transcripts). The output is XML, with a pipeline that can integrate with LLM tooling. It includes a web crawler subsystem with configurable options (depth, pages, user agent, delays, patterns) and alias management for inputs. The tool also supports a Python API exposed as run from onefilellm.
Getting started
Install options:
- Clone and install requirements:
git clone https://github.com/jimmc414/onefilellm.git
cd onefilellm
pip install -r requirements.txt
- Or install as a pip package:
pip install onefilellm
- CLI in editable mode from repo:
pip install -e .
Usage examples:
onefilellm ./docs/ https://github.com/user/project/issues/123
Environment key for GitHub API access:
export GITHUB_TOKEN="your_personal_access_token"
Python API usage:
from onefilellm import run
run(["./docs/"])
Recent releases
- Releases section shows latest 0 with no entries.
Traction
- Stars: 2003
- Forks: 176
- Open issues: 3
Behind the repo
- Repository: jimmc414/onefilellm
- Language: Python
- License: MIT
- Created: 2023-03-17
- Last push: 2026-06-12
- Topics include arxiv, doi, github, ipynb, llm, papers, pdf, pmid, pull-request, repository, sci-hub, text, tiktoken, youtube-transcript-api
Caveats
- License: MIT
- Age: Created 2023-03-17; last push 2026-06-12
- Dependencies include yt-dlp and youtube-transcript-api for YouTube transcripts; troubleshooting guidance includes upgrading these tools via pip.
Notes:
- All install/usage commands were quoted exactly as in the README where provided.