Skip to content
| Marketplace
Sign in
Visual Studio Code>Other>Cite4MeNew to Visual Studio Code? Get it now.
Cite4Me

Cite4Me

Cite4Me

|
9 installs
| (0) | Free
Generate academic citations
Installation
Launch VS Code Quick Open (Ctrl+P), paste the following command, and press enter.
Copied to clipboard
More Info

Cite4Me

Automatically find and generate citations for any statement - right inside VS Code.

Cite4Me uses large language models and academic search engines to find supporting (or refuting) evidence for claims in your text. Select a sentence, run the command, and get back relevant papers or news articles complete with formatted citations ready to paste into your document.

Features

Generate Citations for Any Statement

Select text in your editor (or place your cursor on a sentence) and run Cite4Me: Generate Citations (Ctrl+Shift+C / Cmd+Shift+C). Cite4Me will:

  1. Generate targeted search queries from your statement
  2. Search academic databases or news sources for relevant papers and articles
  3. Classify each result as Support, Refute, or Neutral
  4. Extract the key evidence sentence from each source
  5. Refine your original statement based on the evidence found
  6. Suggest follow-up queries to strengthen the search

Results appear in the sidebar tree view and a webview panel beside your editor.

Scan TeX files for Uncited Claims

Use Scan for Potential Citations to automatically detect sentences in your TeX files that may need a citation. The scanner combines fast heuristic checks (looking for phrases like "studies show", "research indicates", statistics, etc.) with LLM-based analysis for ambiguous cases. Flagged sentences appear as blue wavy underlines - hover to see why, then use Quick Fix (Ctrl+.) to generate a citation directly.

Multiple Citation Formats

Every result includes citations in three formats, each with a one-click copy button:

  • APA
  • Vancouver
  • BibTeX

Search History

All of your past searches are saved and accessible from the Search History view in the sidebar. Click any previous search to restore its full results instantly.

Follow-Up Queries

After each search, Cite4Me suggests 3–5 follow-up queries to help you uncover additional evidence. Click any suggestion to run another search iteration - new results are merged with the existing ones.

Supported Sources

Source Database Best For
Academic Papers OpenAlex
News Articles NewsAPI.ai

Supported LLM Providers

Cite4Me works with multiple LLM providers - bring your own API key:

  • GitHub Models (GPT-4)
  • Anthropic (Claude Sonnet 4.5)
  • Google Gemini (Gemini 3.0 Flash)
  • OpenRouter (GPT-OSS-120B)

Getting Started

  1. Install the extension from the VS Code Marketplace.
  2. Configure API keys - open the Command Palette (Ctrl+Shift+P) and run Cite4Me: Configure API Keys. You need at least one LLM provider key. A NewsAPI.ai key is required only for news article searches.
  3. Select text in your editor and press Ctrl+Shift+C, or right-click and choose Generate Citations from the context menu.
  4. Choose your preferred LLM provider and source type (academic papers or news).
  5. View results in the sidebar and webview panel.

Commands

Command Description
Cite4Me: Generate Citations Generate citations for the selected text or sentence at cursor
Cite4Me: Scan for Potential Citations Scan the current paragraph/selection for sentences that may need citations (LaTeX)
Cite4Me: Configure API Keys Set up API keys for LLM providers and NewsAPI.ai
Cite4Me: Show Results Panel Open the results webview panel
Cite4Me: Clear Results Clear the current citation results
Cite4Me: Clear History Clear all saved search history

How It Works

Cite4Me runs an iterative search → evaluate → refine pipeline:

  1. An LLM generates multiple search queries based on your statement.
  2. Each query is sent to OpenAlex (papers) or NewsAPI.ai (news) to retrieve candidate sources.
  3. The LLM evaluates each source against your statement, classifying it as supporting, refuting, or neutral.
  4. Extended content (citation contexts, abstracts) is fetched from Semantic Scholar when available to improve classification accuracy.
  5. Cross-paper corroboration strengthens evidence where multiple sources agree.
  6. The LLM refines your original statement to be more precise based on the evidence.
  7. Follow-up queries are generated for additional search iterations.

Requirements

  • VS Code 1.84.0 or later
  • At least one LLM provider API key (GitHub Models, Anthropic, Google Gemini, or OpenRouter)
  • NewsAPI.ai API key (optional, only needed for news article searches)
  • Contact us
  • Jobs
  • Privacy
  • Manage cookies
  • Terms of use
  • Trademarks
  • Your Privacy Choices
  • Consumer Health Privacy
© 2026 Microsoft