// Unified poster and standalone image generation skill with hard routing. Produces one or more final poster/image assets as requested, using mandatory HTML-based layout for information-heavy posters and image-first generation only for strictly visual assets.
| name | poster-designer |
| description | Unified poster and standalone image generation skill with hard routing. Produces one or more final poster/image assets as requested, using mandatory HTML-based layout for information-heavy posters and image-first generation only for strictly visual assets. |
Poster Designer is a unified poster generation skill that outputs the requested number of final poster/image assets, while internally selecting the appropriate generation strategy.
Unlike soft, heuristic-based tools, this skill enforces hard routing rules to ensure that information-heavy posters are generated using deterministic layout rather than direct image generation.
The skill supports two internal strategies:
Strategy selection is final and non-negotiable once routing conditions are met.
Use this skill whenever the desired output is:
Typical use cases include:
This skill always produces:
The skill must never output:
Only the final image is returned.
Poster Designer applies mandatory routing rules when selecting the generation strategy.
These rules override upstream planner preferences and must not be bypassed.
The Layout-first (HTML-based) strategy MUST be used if any of the following conditions are satisfied:
The poster is described as an announcement, notice, or formal milestone
The poster is intended for long-term display, including but not limited to:
The request explicitly mentions or implies:
The poster contains more than one semantic text role, such as:
When any of these conditions are met:
When using browser screenshot tools, the viewport width MUST be explicitly set to match the layout width (default: 1080px), and the output filename MUST clearly indicate the poster's purpose.
This strategy prioritizes:
It is the default strategy for any poster that functions as an information artifact rather than a purely visual symbol.
The Image-first strategy is ONLY permitted when all of the following conditions are satisfied:
If any of these conditions are not met, Image-first generation must not be used.
Regardless of strategy, Poster Designer enforces the following principles:
The poster must clearly communicate:
Poster Designer introduces domain-specific visual anchors instead of generic backgrounds.
Examples include:
Purely abstract backgrounds without semantic meaning are insufficient.
Before final screenshot, verify:
No critical content is obscured
Layer discipline
Visual hierarchy
Text-on-image safety
If any of these checks fail, the poster must be adjusted before taking the final screenshot.
A successful output must be suitable for:
If the output resembles:
Then the skill has failed.
"Create a formal milestone announcement poster for Xinference celebrating 5,000,000 Docker Hub downloads. The poster must include a title, a prominent numeric highlight, explanatory text, acknowledgements, and footer information suitable for long-term display on README and official websites."
Poster Designer is an intent-aware, rule-enforced poster generation skill.
It guarantees that information-heavy posters are generated using deterministic layout, while reserving direct image generation strictly for minimal, visual-only use cases.
Generate and edit PowerPoint presentations (.pptx). Use for: creating slide decks, pitch decks, or presentations from scratch; reading, parsing, or extracting content from .pptx files; editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger only when the user explicitly asks for a deck, slides, presentation, PPT/PPTX, or references a .pptx filename. Do not use for standalone posters, images, banners, illustrations, or social graphics unless the requested deliverable is a presentation file.
Expert AI Agent Builder skill. Helps users dynamically create, configure, and build custom AI agents with specific capabilities, knowledge bases, and tools.
Evidence-first retrieval-augmented reasoning skill for decision-critical scenarios. Retrieves information from a specified knowledge base, extracts verifiable evidence, detects conflicting claims, and evaluates answer sufficiency with explicit confidence and risk signals. Produces traceable outputs suitable for agent-level decision control and escalation.
