- Home
- Skills
- Agents & Orchestration
- make natural academic slides
make natural academic slides
Automated medical academic PPTX generator with high-res PDF asset extraction and evidence traceability.
Free
Works with the AI Agents you already use
About this skill
Academic and medical presentations often suffer from a disconnect between the source literature, the visual slides, and the speaker script. This skill bridges that gap by transforming raw research papers, DOIs, or existing drafts into professional, editable PPTX presentations with built-in evidence traceability and high-fidelity source material integration.
What it does
- Automated Literature Synthesis: Uses supplied PDFs or retrieves accessible full-text PDFs via DOI or title and extracts key sections including background, methods, results, and discussion.
- High-Fidelity Asset Extraction: Extracts embedded PDF images at native resolution and captures vector figures, flowcharts, and tables at a default 600 DPI.
- Traceable Narrative Design: Generates a synchronized
deck.jsonthat records slide claims and their source locations for review. - Interactive Review Platform: Produces a standalone HTML dashboard for offline review, featuring slide previews, speaker notes, and evidence panels.
- Template Adaptation: Rebrands existing content or new research into custom corporate or university PPTX/POTX templates while preserving layout logic.
- Speaker Scripting: Drafts contextual scripts for every slide, focusing on research highlights, novelty, and clinical significance.
How it works
- Ingest: Provide a PDF, a DOI, or an existing PPTX file to serve as the baseline for the report.
- Plan: The skill analyzes the paper's highlights and creates a
focus_planto allocate slide count based on study impact. - Execute: It runs Python and Node.js scripts to extract high-resolution assets and generate the editable PPTX structure.
- Review: Open the generated
index.htmlto audit the evidence, add feedback, or select specific discussion directions.
Frameworks & tools
The system uses Python 3.10+ for data orchestration and Node.js 18+ for slide rendering. It integrates with @oai/artifact-tool for PPTX manipulation and Pillow for PDF image processing.
Evidence review
The workflow links slide content to source passages, preserves original figure labels, and provides checks for consistency between slides, notes, and evidence records.
Use cases
- Journal Clubs: Quick conversion of a new NEJM or Lancet paper into a structured literature report for peers.
- Research Defense: Building a traceable slide deck for thesis or grant defense where every claim must be backed by data.
- Clinical Teaching: Creating case-based teaching slides with direct links to clinical guidelines and evidence.
- Template Migration: Moving an old research presentation into a new department-branded template without manual re-typing.
Known limitations
Requires a local environment with Node.js and Python to run the rendering and validation scripts. Full PDF access is necessary for high-resolution image extraction.
Helps extract, organize, and check key data from research papers. A final human review is recommended before use or publication.
Real example
The GitHub repository includes a published 17-slide Chinese demonstration, “膝骨关节炎:用疼痛重新审视影像分级” (Knee osteoarthritis: re-examining imaging grades through pain), version v1.0-beamer-demo. It presents the paper “An algorithmic approach to reducing unexplained pain disparities in underserved populations.”
The deck covers the research background, pain-focused learning objective, imaging inputs and joint training, disparity analysis, validation and ablation studies, results, discussion, strengths and limitations. Its review page displays slide previews, page-by-page speaker notes, evidence records and feedback controls, with an option to export a revision checklist.
View the existing 17-slide demonstration · GitHub repository
Requires Node.js 18+ and Python 3.10+. High-resolution asset extraction depends on having the source PDF file available. PPTX rendering requires the @oai/artifact-tool.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 5 days ago
- Free with an Agensi account
Requires
Permissions
Allowed Hosts