| name | poster-design |
| description | Design polished, print-ready scientific posters using Python and/or frontend web tools. Use this skill when the user asks for a conference poster, research poster, academic poster, or large-format print layout. |
Poster Design
Use this skill for scientific, academic, conference, or large-format research posters.
Default Deliverables
Unless the user asks otherwise, produce a short design brief, editable source files, supporting figure assets, and a print-ready PDF under IDEA's standard output directory: /app/static/{user_id}/{session_id}.
Workflow
- Establish or infer the poster size, orientation, audience, constraints, and desired density. Design at final print size from the start.
- Create a concise design brief covering the poster goal, core takeaway, visual strategy, layout strategy, and production method.
- Triage content aggressively: prioritize the main finding, one or two dominant figures, short methods context, implications, and minimal references.
- Prefer Python for data graphics, maps, QR codes, and analysis-driven assets. Export figures as SVG/PDF when possible, or high-resolution PNG when raster is required.
- Prefer HTML/CSS for final poster composition unless the user requests Python-only. Use fixed physical dimensions, CSS Grid/Flexbox, print rules, zero browser margins, 100% scale, and background graphics enabled.
- Export a single-page PDF and verify the page size matches the target dimensions.
Design Defaults
Use a clear grid, disciplined alignment, readable poster-scale typography, restrained color, and one or two strong focal areas. Avoid manuscript-length text, low-resolution screenshots, tiny figures, too many colors or fonts, and heavy boxed layouts unless the user asks for that style.
Preflight
Before delivery, inspect the PDF for exact page size, page count, bounds, overlap, font rendering, image sharpness, vector crispness, contrast, scannable QR codes, and a clear 10-second takeaway.
