Literature Review Synthesizer – turn your reading notes into structured lit review outputs (BYOK, fully local)

Share & showcase
1

Disclaimer

Is this project open source? Yes
Is this project completely free? No (free tier + one-time Pro license at $19.99)
Is this project vibe-coded beyond the author’s ability to comprehend how it works? No
Community Directory: Literature Review Synthesizer - Obsidian Plugin

Hi all,

I built this because I kept running into the same wall. You read 80 papers, take notes in Obsidian, and then the synthesis stage – the part where you actually figure out what the field is doing – takes weeks of manual work. Tools like Elicit are useful but they live outside your vault, and if you care about keeping your notes local, they’re a non-starter.

What it does

You select a folder or tag group of your reading notes, pick a synthesis mode, and the plugin sends your notes to your own LLM (via your own API key) and writes the result back into your vault as a new note with frontmatter and backlinks.

Four modes in v1.0:

  • Thematic Synthesis – common themes and findings across your sources
  • Methodological Comparison Matrix – side-by-side breakdown of study designs, samples, measures
  • Research Gap Analysis – where the literature is thin, contradictory, or silent
  • Draft Literature Review Section – a proper academic draft you can actually edit and cite from

Supports OpenAI and Anthropic out of the box. Also works with any OpenAI-compatible endpoint (OpenRouter, local Ollama, etc.) via custom base URL.

Your notes and API key never leave your machine. No server, no account, no subscription.

Free tier and Pro

The free tier gives you Thematic Synthesis with 3 runs per month. Pro (one-time purchase, $19.99) unlocks everything with no usage limits.

How to install

Search “Literature Review Synthesizer” in Obsidian’s Community Plugins browser. Settings → Community plugins → Browse → search the name → Install → Enable.

Then add your API key in the plugin settings and you’re set.

Links

Happy to answer questions. Feedback on the synthesis output quality especially welcome – I’m tuning the prompts based on real use.