- Home
- Skills
- Agents & Orchestration
- academic presenting
academic presenting
Builds and audits scientifically rigorous academic presentations with automated evidence tracing and quality reports.
Free
Works with the AI Agents you already use
academic presenting
Example session with this skill installed
I have notes from my latest clinical trial on drug X. Help me set up the intake for a 15-minute research talk at the Cardiology Summit.
- Read your context and instructions
- Compiled the academic presenting
I have initiated the intake process in local/cardiology-summit-talk/intake.md. I have mapped your trial notes to the core requirements and identified four specific areas needing clarification to complete the presentation specification.
Intake Progress: Cardiology Summit Talk
The following table summarizes the current state of your presentation intake:
| Category | Status | Data Point |
|---|---|---|
| Occasion | Complete | Cardiology Summit, 15-minute slot |
| Presenters | Incomplete | Lead author identified; co-presenters needed |
| Title Slide | Pending | Needs exact occasion line |
| Core Claims | Partial | Drug X efficacy mapped; safety data locators needed |
| References | Pending | Defaulting to full references slide |
| Evidence Status | Warning | 3 claims marked as needs-expert-review |
I have identified the following open questions. Please answer these to proceed to the Evidence Integrity phase:
- Who are the co-presenters and their specific affiliations for the title slide?
- Do you have the specific p-values for the secondary endpoints, or should I flag these for manual entry?
- Is there a specific institutional template or color palette required for the Cardiology Summit?
- Are there specific figures from the trial manuscript you have permission to reuse, or should I generate new charts from the raw numbers?
Next steps
- Provide the co-presenter names and affiliations.
- Confirm the source locators for the secondary endpoint p-values.
- Once answers are provided, I will run
npm run validateon the draft spec.
Connects securely to your tools. The creator never sees your data.
About this skill
Developing a high-stakes scientific or clinical presentation requires more than just a template. It requires rigorous evidence tracking and structural integrity. This skill builds and reviews academic presentations by enforcing a strict storyboard-first workflow and automated quality checks.
What it does
- Evidence verification validates scientific claims by recomputing derived numbers and mapping every statement to specific source locators.
- Structured intake conducts a targeted nine-point interview to capture the occasion, co-presenters, and core research notes.
- Storyboard generation creates a comprehensive slide-by-slide plan that must be approved before any rendering occurs.
- Automated quality reports flags untraceable numbers, needs-review items, and internal inconsistencies before completion.
- PowerPoint rendering outputs editable
.pptxfiles based on validated specifications, ensuring layout changes never compromise data accuracy.
How it works
- Intake where the agent maps notes to a structured
intake.mdand clarified open questions. - Analysis of source material where claims are assigned states like
mechanically-validatedorneeds-expert-review. - Spec writing via
schemas/presentation.schema.jsonto define claims, caveats, and source references. - Approval gate where you review a generated
storyboard.mdbefore the agent builds the final deck. - Quality report generation that hands over the deck, a contact sheet, and a list of items requiring expert eyes.
Frameworks & tools
This skill utilizes a Node.js-based toolchain involving storyboard.mjs for rendering and npm scripts for validation. It generates standard PowerPoint .pptx files and uses Markdown-based schemas for intermediate content representation.
Why this beats prompting it yourself
Standard prompts often hallucinate citations or subtly alter scientific meanings to fit a narrative. This skill enforces a schema-based intermediate layer that separates claims from layout, ensuring data integrity is maintained throughout the design process.
Use cases
- Research talks where every claim must be backed by a specific line in a published paper.
- Journal clubs that require re-computation of clinical trial percentages and CI plausibility.
- Teaching materials needing a structured review of funding contexts and team credits.
- Slide upgrades where existing decks are diagnosed for structural weaknesses and rebuilt from a verified spec.
Known limitations
Does not support editing existing PowerPoint files in place. Cannot perform actual scientific verification, only mechanical validation and source tracing. HTML output via Slidev is planned but not currently available.
Requires manual approval of storyboards. Cannot edit existing .pptx files in place. Limited to mechanical validation of provided data; cannot verify external truth.
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 today
- Free with an Agensi account
Requires