I turned my open-source paper-reading CLI into a desktop app — and gave it an honesty rule

Описание и применение

PaperLens Pro преобразует CLI‑инструмент для чтения научных статей в настольное приложение с поддержкой пакетной обработки, предпросмотра и экспорта, сохраняя правило честности, при котором модель указывает «неясно», если информация отсутствует.

Задача

Users needed a more efficient way to read and summarize multiple research papers without generating fabricated information.

Результат

PaperLens Pro enables efficient, honest summarization of multiple research papers with batch processing, preview, and export features, while preventing fabricated content through an explicit "unclear" rule.

Рабочий процесс

• Install the source package and run the double‑click launcher.
• Add PDFs via drag‑and‑drop, folder selection, or individual files to the queue.
• The app extracts text (pypdf), truncates to 12,000 characters if needed, and sends it to the chosen LLM backend with a strict prompt.
• The LLM returns a Markdown report with five sections: TL;DR, key contributions, method, key results, reproduction notes.
• The app displays the report in‑app and allows one‑click export to a single .md per paper or a combined file for Obsidian, Notion, or a lab wiki.
• If the PDF has no extractable text, the app informs the user that OCR is required.
• The user can configure API keys for Ollama, Groq, or any OpenAI‑compatible endpoint in a local settings file.
• The app respects the honesty rule: if the paper text does not cover a requested detail, the report contains "unclear" instead of fabricated content.
• The user can tweak the prompt or build a standalone executable with PyInstaller if desired.

Доказательства

Every week, my "to read" folder gained ten PDFs and lost none. The classic fixes didn’t work for me: pasting papers into a chat window one at a time is slow, and the summaries I got back had a worse problem — they sounded confident about things the paper never said.

Источник

The CLI worked. But it only worked for people comfortable in a terminal, one paper at a time, with no way to see a queue, preview a report, or export a whole batch into notes.

Источник

The one design decision I care most about is in the prompt. It instructs the model: if the paper’s text doesn’t cover something, write "unclear" — do not fill the gap with something plausible.

Источник

PaperLens Pro is delivered as a source package with double-click launchers, not a signed installer.

Источник