Repository navigation
TextFold
Long papers, web pages, and AI replies are often too dense to skim and too important to skip. TextFold folds a document into five levels of detail, from a one-line topic map down to the full original text, so a reader can zoom in only where needed.
TextFold accepts arXiv, PubMed Central, bioRxiv, medRxiv, Wikipedia, GitHub READMEs, web pages, PDFs, Office files, and EPUBs (README). Readers can view the original PDF beside the text, highlight and annotate, and ask questions about it. It integrates with a local Zotero library, supports translation, and exports pages as standalone HTML.
It ships as a desktop app for macOS and Windows and as a Python package (pip install textfold, then python3 -m textfold serve). It uses a model the user already has, such as Claude Code, Codex, or Pi, or a ChatGPT plan, OpenRouter, or an API key. Plugins for Claude Code and Pi fold each agent reply into the same five levels, and a CLI returns a document one level at a time for scripts.
The repository was created on 2026-10-04 under the Apache-2.0 license and was formerly named PaperFold. At the time of review it had about 401 stars and 4 open issues.
TextFold is placed in Assess within Tools. It addresses information overload in a reusable way and integrates with agent workflows, but the project is only weeks old. No first-person production use has been confirmed, so the ring remains Assess.
