telerik/ai-plugins/prompt-enrichment

Claude Code and agent skill by telerik/ai-plugins

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Expand vague or minimal UI requests into rich, specific design briefs before generating Kendo components. Trigger whenever a user asks to build ANY kind of user interface using Kendo, and the request lacks specific layout, data, or component details — dashboards, apps, forms, portals, landing pages, e-commerce views, social feeds, chat interfaces, settings screens, onboarding flows, or any other UI. Skip when the user already provides a detailed spec. This skill decides WHAT to build; the Kendo design/styling skill decides HOW to style it.

Install

npx vskill@latest install telerik/ai-plugins/prompt-enrichment

SKILL.md

Prompt Enrichment

Short or generic UI prompts produce bland, generic layouts with placeholder data. This skill expands any vague Kendo UI request into a detailed design brief before code is written.

Workflow Contract

After reading this skill, stop and execute the Decision Table before doing any research, tool execution or implementation work.

  1. Choose the matching Decision Table row.
  2. Read the reference file named by that row before producing a brief, scaffold, clarification question, or skip decision.
  3. If the selected row is Clarification triage, ask the clarification question and wait for the user's answer before returning to this workflow.
  4. Do not create a design brief from this file alone. The enrichment gate is incomplete until the selected reference workflow has been followed.
  5. Only after this workflow is complete may the orchestrator continue with accessibility, component, layout, theming, icon, validation, or project-file work.

Decision Table

This is the single source of truth for picking a pattern. Each row lists the trigger signals, the deliverable, the pattern name, and the reference to load for full instructions. Reference files do not repeat trigger criteria — they are loaded only after the pattern has been picked here.

Pick this pattern when...The deliverable is...PatternReference
The prompt is vague or generic (under ~30 words, no layout/data/component specifics, generic terms like "dashboard", "page", "app"). The user wants a UI but hasn't said what's in it.A detailed design brief: user role, information architecture, spatial layout with dimensions, Kendo component plan, realistic sample data, density/tone, and interaction spec.Single-view enrichmentreferences/enrichment-dimensions.md
The prompt targets a multi-view application (signals: "app", "system", "platform", "portal", multiple features, workflows spanning screens). Multiple screens need shared architecture.An app scaffold (identity, roles, view map, navigation, data model, visual identity) + a design brief for the first view.App scaffoldingreferences/app-scaffolding.md
The prompt contains possessive references ("our team", "my data") or named entities without context ("for Acme Corp"). Real context exists but wasn't stated.1–2 targeted clarifying questions with tappable options + escape hatch, followed by enrichment or scaffolding.Clarification triagereferences/clarification-triage.md
The user already provides a detailed spec — specific components, data fields, layout instructions, and content. Nothing is vague.No enrichment needed. Build directly.Skip—
You need to select a spatial layout pattern for any view (during enrichment or scaffolding).A specific layout pattern with spatial structure, Kendo components, and "when NOT to use" guidance.Layout selectionreferences/layout-patterns.md

If a prompt does not clearly match any row, default to single-view enrichment — it's the safest starting point. If both row 1 and row 2 match (vague prompt + app scope), run app scaffolding first, then enrich the primary view.

Core Principles

  1. Specificity produces better UIs. A concrete user role + realistic data + explicit spatial layout always outperforms a generic prompt — even when the scenario is fictional.
  2. Enrich any UI type. This skill covers dashboards, e-commerce pages, chat interfaces, settings screens, onboarding wizards, social feeds, media galleries, landing pages — not only business dashboards.
  3. Layout must be spatially explicit. Never just name a pattern ("List-Detail"). Describe panel widths, what's above/below the fold, column/row structure, fixed vs. scrollable regions.
  4. Data must be realistic. Never use "[placeholder]" or "Data 1". Invent plausible names, amounts, dates, and cross-component relationships.
  5. Don't duplicate implementation. Defer to Kendo Agentic UI Generator assistants (#kendo_component_assistant, #kendo_layout_assistant, etc.) for API usage, code, and theming. This skill produces the design brief they consume.

Brief Format

Every enrichment — whether single-view or per-view within an app — produces a brief in this shape:

## Design Brief: [Descriptive Title]
**User**: [Role + context, 1–2 sentences]
**Primary Task**: [What they do on this screen]
**Layout**: [Pattern name + spatial description with dimensions/ratios]
**Density**: [High/Medium/Low] | **Tone**: [Descriptor]
### Component Plan
1. [Name] — [Kendo component] — [What it shows, with sample values]
2. ...
### Key Interactions
- [What happens on click/filter/selection]
### Sample Data
[Field names, types, 3–5 realistic rows per component]

The brief stays internal unless the user asks to see it. Present the plan conversationally, get confirmation, then build.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/telerik/ai-plugins/prompt-enrichment/badge)](https://verified-skill.com/skills/telerik/ai-plugins/prompt-enrichment)

Something wrong with this skill or its page? Tell us.