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.
How this skill is triggered — by the user, by Claude, or both
Slash command
/kendo-angular-mcp-plugin:prompt-enrichmentThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
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.
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.
After reading this skill, stop and execute the Decision Table before doing any research, tool execution or implementation work.
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... | Pattern | Reference |
|---|---|---|---|
| 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 enrichment | references/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 scaffolding | references/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 triage | references/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 selection | references/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.
#kendo_component_assistant, #kendo_layout_assistant, etc.) for API usage, code, and theming. This skill produces the design brief they consume.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.
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub telerik/ai-plugins --plugin kendo-angular-plugin