| name | science-communication |
| description | Translates academic research into audience-tailored formats including lay summaries, policy briefs, press releases, social media threads, and visual abstracts. |
| version | 1.0.0 |
| tags | ["academic","communication","writing"] |
Science Communication Skill
Overview
This skill transforms academic research into audience-appropriate communication formats. It maintains scientific accuracy while adapting language, structure, depth, and emphasis for different readerships. The skill supports five output formats, each with its own tone, structure, and style guidelines.
Supported Output Formats
| Format | Audience | Length | Tone | Key Elements |
|---|
| Lay Summary | General public | 300-500 words | Accessible, engaging | Why it matters, what was found, in plain language |
| Policy Brief | Decision-makers | 1-2 pages | Authoritative, actionable | Problem, evidence, options, recommendation |
| Press Release | Journalists/media | 400-600 words | Newsworthy, quotable | Hook, key finding, quote, context |
| Twitter/X Thread | Academic Twitter | 5-8 tweets | Conversational, punchy | One key insight per tweet, visual, call to action |
| Visual Abstract | All audiences | 1 image + caption | Visual-first | Core concept, key result, take-home message |
When to Use This Skill
Use when:
- User provides a paper and says "summarize for the public", "explain this simply", "make this accessible"
- User says "write a press release", "prepare a media statement", "draft a news article"
- User wants to share research on social media: "create a Twitter thread", "write a LinkedIn post"
- User says "policy brief", "briefing note", "executive summary for policymakers"
- User wants a visual abstract or graphical summary
- The
academic-paper-review skill has been run and user wants to disseminate findings
- User is preparing a talk or presentation for a non-specialist audience
Workflow
Phase 0: Input
This skill works best when you already have a paper review or decomposition available. If the user hasn't run academic-paper-review or paper-decomposition, do a rapid read of the paper (abstract + introduction + results + conclusion) to extract:
- Core research question
- Key finding / main result
- Significance / why it matters
- One surprising or counterintuitive aspect
- Limitations (mentioned honestly)
Phase 1: Audience Adaptation
For Lay Audience (Lay Summary)
Principles:
- Start with WHY: Lead with the problem or question people care about, not the method
- No jargon: Replace technical terms with everyday analogies. If you must use a technical term, define it in one sentence
- Concrete over abstract: Use specific examples, not general principles
- Human interest: Connect findings to people's lives
Structure:
[Hook: relatable problem or surprising fact]
[Context: what we already knew, in simple terms]
[The New Finding: what the researchers discovered]
[Why It Matters: implications for everyday life]
[Caveat: one honest limitation, stated plainly]
For Policy Brief
Principles:
- Problem-first: Policy makers need to know what problem this addresses
- Evidence-based options: Present 2-3 policy options grounded in the research
- Actionable recommendations: Be specific about what should be done
- Consider feasibility: Acknowledge implementation challenges
Structure:
EXECUTIVE SUMMARY (3 bullet points)
1. PROBLEM STATEMENT
- What is the issue? Who is affected? What is the scale?
2. RESEARCH FINDINGS
- What does the evidence show? How strong is the evidence?
3. POLICY OPTIONS
- Option A: [description, pros, cons, feasibility]
- Option B: [description, pros, cons, feasibility]
4. RECOMMENDATION
- Preferred option with justification
5. LIMITATIONS AND UNCERTAINTIES
- What we don't know, caveats, need for further research
For Press Release
Principles:
- Newsworthiness: Why should anyone care? What's new?
- Inverted pyramid: Most important info first, details later
- Quote-worthy: Include at least one quotable statement from "the researchers"
- No hype: Avoid overclaiming. Use measured language
Structure:
HEADLINE: [Active voice, key finding, < 15 words]
DATELINE — [Lead paragraph: who, what, when, where, why in 1-2 sentences]
[Quote from lead researcher — authentic, not marketing-speak]
[2-3 paragraphs of supporting detail — methodology, context, significance]
[Boilerplate: about the institution / funding source]
Media Contact: [name, email]
Paper Reference: [citation or DOI]
For Twitter/X Thread
Principles:
- One idea per tweet: Each tweet is self-contained and valuable
- First tweet is the hook: Make people want to click "Show more"
- Use numbers and specifics: "94.2% accuracy" not "highly accurate"
- Include visuals: Charts, diagrams, or paper figures (with attribution)
- End with CTA: Link to paper, invite discussion
Structure:
Tweet 1 (Hook): Surprising finding or counterintuitive result
Tweet 2: What was previously known / the problem
Tweet 3: What the researchers did (1 sentence)
Tweet 4: Key result with number
Tweet 5: Why this matters / what changes
Tweet 6: One limitation or open question
Tweet 7 (optional): Link to paper + "What do you think?"
For Visual Abstract
Principles:
- One core concept: Don't try to show everything
- Left-to-right or top-to-bottom flow: Guide the eye naturally
- Minimal text: Labels only, no paragraphs
- Color coding: Consistent colors for consistent concepts
Generate a text description that the user can give to a designer or use with image-generation:
Visual Abstract Description:
- Layout: Horizontal, 3 panels left to right
- Panel 1 (Problem): [description of visual element]
- Panel 2 (Method): [description of visual element]
- Panel 3 (Result): [description of visual element]
- Color scheme: [suggested colors and their meanings]
- Take-home message: [1 sentence to place at bottom]
Phase 2: Generate Chosen Format(s)
Ask the user which format(s) they want, or infer from context. Generate one or more formats.
Phase 3: Quality Review
Before finalizing, check:
Lay Summary Checklist
Policy Brief Checklist
Press Release Checklist
Twitter Thread Checklist
Integration with Other Skills
- After
academic-paper-review: Transform the detailed review into accessible formats
- After
paper-decomposition: Use the structured extraction to identify the most communicable insights
- Before
podcast-generation: Use the lay summary as script foundation
- Before
image-generation: Use the visual abstract description as an image prompt
Quality Checklist