All products

Build a research presentation with AI

Turn a paper or project folder into a browser presentation. Trace claims and visuals to their sources and keep extra methods slides ready for questions.

Build a research presentation with AI when the talk needs a clear argument and every comparison must survive questions from the audience. The research-presentation skill starts from a paper, a research directory or an existing deck alongside its sources. It maps evidence before deciding what each slide should establish.

A chart copied from a PDF can support a discussion of the published figure. It does not give you the underlying observations. Retaining the page and figure number lets the audience trace the image while you reserve claims about individual measurements for cases where the data are available.

What it does

  • Plan a sequence around the audience and time available.
  • Tie numerical or comparative statements to a specific source.
  • Keep exploratory outcomes visibly qualified on the slide.
  • Move detailed methods or alternative comparisons into backup slides.
  • Open the finished HTML locally to check layout, navigation and source references.

Install and use

  1. Install the skill in your agent, provide authorised sources and make Python 3 available for scaffolding and validation. Use a local browser for visual review.
  2. Ask for a paper walkthrough that shows one main result, with extra slides on the evaluation setup.
  3. Check each figure and visible number against the accompanying evidence record before presenting.

FAQ

Source

Made by DELONG-L. Original skill: research-presentation, published on GitHub under the MIT licence.

View on GitHub