agentleFS
yuzhaopeng-upyuzhaopeng-up/teleagent-skills/skills/visualization-renderer/SKILL.md

visualization-renderer

Skill1 starsChanged 5 months agoApache-2.0View on GitHub

Visualization Renderer — converts structured data into interactive ECharts charts and dashboards. 4-Phase forced orchestration: data feature analysis -> ECharts config generation -> HTML page rendering -> dashboard multi-chart layout. Depends on diagram-drawing and web-artifacts-builder components, orchestrated via Phase-Orchestrator. Triggers: chart display, visualization, bar chart, line chart, pie chart, dashboard, kanban, data display.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Style: use Phase 1 warm theme colors, chart container width:100%/height:400px, card-style rounded 8px shadow, title 18px bold body 14px

    line 175
  2. Distinctive:

    Condition: Only executed when chart_type == "dashboard".

    line 194

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Visualization Renderer

Core visualization component. Converts structured data into interactive charts. Triggered when output_format is chart/dashboard.

Parameters

ParameterTypeRequiredDefaultDescription
dataJSONYes-Aggregated data
chart_typeOptionNoautoauto/bar/line/pie/table/scatter/radar/dashboard
titleTextNoData Query ResultChart title
themeOptionNowarmwarm/cool/dark/light
interactiveSwitchNoonTooltip, zoom, filter

Chart Recommendation Rules

Data FeatureRecommended ChartReason
Single dimension + single metric + rankingHorizontal barIntuitive comparison
Single dimension + single metric + proportionPieShow proportional relationship
Time series + trendLineShow change trend
Multi dimension + multi metricGrouped barCompare multiple dimensions
Multi metric correlationScatterShow correlation
Comprehensive displayDashboardMulti-angle display

Supported Chart Types

TypeECharts typeUse CaseData Requirement
barbarRanking, comparison1 dimension + 1-3 metrics
linelineTrend, time seriesTime dimension + 1-3 metrics
piepieProportion, composition1 dimension + 1 metric
scatterscatterCorrelation, distribution2 numeric metrics
radarradarMulti-dimensional evaluation3-8 metrics
tabletableDetailed dataAny structure
dashboardmixComprehensive dashboardMultiple data groups

4-Phase Forced Orchestration Protocol

Orchestrated via Phase-Orchestrator. Each Phase launches a general-type sub-Agent via task tool.

Phase 1: Data Feature Analysis and Chart Recommendation

Analyze data features and recommend chart type.

Execution prompt (injected into sub-Agent):

You are the Phase 1 executor of the Visualization Renderer. Analyze data features and recommend chart type.

Step 1: Data feature extraction

  • Dimension count: 1/2/multiple/none
  • Dimension type: categorical/temporal/numeric
  • Metric count: 1/2/multiple
  • Metric type: count/amount/ratio/duration
  • Row count: small (<10)/medium (10-50)/large (>50)
  • Special features: has_ranking/has_trend/has_proportion/has_comparison

Step 2: Chart recommendation (when chart_type=auto)

Auto-select based on recommendation rules table. When multiple charts fit, generate recommendation list with first as optimal choice, annotated with reason.

Step 3: Color scheme

  • warm: #FF6B35, #F7C59F, #EFEFD0, #004E89, #1A659E
  • cool: #2196F3, #00BCD4, #4CAF50, #FF9800, #9C27B0
  • dark: dark background + bright colors
  • light: light background + saturated colors

Step 4: Interaction suggestions

  • Data > 10 rows -> enable dataZoom
  • Multi dimension -> enable legend filter
  • Has outliers -> add markPoint

Output JSON:

{
  "chart_recommendation": {
    "primary": "bar",
    "alternatives": ["pie", "table"],
    "reason": "Single dimension ranking comparison, bar chart is most intuitive"
  },
  "data_features": {
    "dimension_count": 1,
    "dimension_type": "category",
    "metric_count": 1,
    "metric_type": "count",
    "row_count": 5,
    "has_ranking": true,
    "has_trend": false,
    "has_proportion": true
  },
  "color_scheme": ["#FF6B35", "#F7C59F", "#004E89", "#1A659E", "#EFEFD0"],
  "interaction_config": {
    "tooltip": true,
    "legend": true,
    "dataZoom": false,
    "markPoint": true
  },
  "phase_status": "success"
}

Phase 2: ECharts Configuration Generation

Generate ECharts configuration JSON based on Phase 1 results.

Execution prompt (injected into sub-Agent):

You are the Phase 2 executor of the Visualization Renderer. Generate ECharts config based on Phase 1 analysis results.

Base configuration structure:

{
  "title": {"text": "<title>", "left": "center"},
  "tooltip": {"trigger": "axis|item"},
  "legend": {"data": ["<metric_name>"], "bottom": 10},
  "xAxis": {"type": "category", "data": ["<dimension_value_list>"]},
  "yAxis": {"type": "value", "name": "<metric_unit>"},
  "series": [{"name": "<metric_name>", "type": "bar|line|pie", "data": [numeric_list]}]
}

Chart-specific configuration:

  • bar: horizontal bar swaps xAxis/yAxis; grouped bar uses multiple series; stacked bar adds stack attribute; add label={show:true, position:"top"}
  • line: smooth:true for curve; areaStyle:{opacity:0.3} for area chart; markPoint/markLine for data markers; time axis xAxis type:"time"
  • pie: radius:["40%","70%"] for donut chart; labelLine for label guide lines; roseType:"area" for Nightingale chart
  • Theme colors: color:["<Phase 1 color scheme>"]
  • Interaction: dataZoom:[{type:"slider"}]; toolbox:{feature:{saveAsImage:{}, dataView:{readOnly:true}}}

Output JSON:

{
  "echarts_option": { "...full ECharts configuration" },
  "chart_meta": {
    "chart_type": "bar",
    "dimensions": ["plan_type"],
    "metrics": ["ticket_count"],
    "data_points": 5
  },
  "phase_status": "success"
}

Phase 3: HTML Page Generation

Load component: skill -> web-artifacts-builder

Generate a complete HTML page containing the ECharts chart.

Execution prompt (injected into sub-Agent):

You are the Phase 3 executor of the Visualization Renderer. Load web-artifacts-builder skill and generate an HTML page containing the ECharts chart.

Page requirements:

  • Structure: header(title + query summary) -> main(ECharts chart container) -> footer(data source + generation time)
  • Tech stack: ECharts CDN https://cdn.jsdelivr.net/npm/echarts@5/dist/echarts.min.js, responsive layout, vanilla CSS
  • Style: use Phase 1 warm theme colors, chart container width:100%/height:400px, card-style rounded 8px shadow, title 18px bold body 14px
  • Interaction: legend toggle data series, hover tooltip, toolbox save image, window resize auto-fit
  • Phase 2 echarts_option embedded in JavaScript initialization code

Output JSON:

{
  "html_content": "<complete HTML string>",
  "page_meta": {
    "title": "<chart title>",
    "chart_count": 1,
    "responsive": true
  },
  "phase_status": "success"
}

Phase 4: Dashboard Multi-Chart Layout (dashboard mode only)

Condition: Only executed when chart_type == "dashboard".

Execution prompt (injected into sub-Agent):

You are the Phase 4 executor of the Visualization Renderer. Layout dashboard multi-chart arrangement.

Layout rules:

  • Grid: 2 columns (3 columns on large screen) auto-fit
  • Top-left: primary metric bar/pie chart
  • Top-right: trend line chart
  • Bottom-left: TOP10 horizontal bar ranking
  • Bottom-right: data detail table
  • Unified title bar (title + time range selector), unified theme colors, global filter (region/time/plan type)
  • Each chart has independent container and echarts_option, linkage: click one chart dimension -> other charts filter accordingly

Output JSON:

{
  "dashboard_html": "<complete Dashboard HTML string>",
  "dashboard_meta": {
    "chart_count": 4,
    "layout": "2x2",
    "charts": [
      {"type": "bar", "title": "...", "position": "top-left"},
      {"type": "line", "title": "...", "position": "top-right"},
      {"type": "pie", "title": "...", "position": "bottom-left"},
      {"type": "table", "title": "...", "position": "bottom-right"}
    ]
  },
  "phase_status": "success"
}

Phase Data Passing

  • Phase 1 -> 2: { "chart_recommendation": {...}, "color_scheme": [...], "interaction_config": {...} }
  • Phase 2 -> 3: { "echarts_option": {...}, "chart_meta": {...} }
  • Phase 3 -> 4: { "html_content": "...", "page_meta": {...} }

Degradation Strategy

PhaseFailure Handling
Phase 1Recommendation fails -> default to bar chart
Phase 2Config generation fails -> degrade to Markdown table
Phase 3HTML generation fails -> degrade to plain text data display
Phase 4Dashboard fails -> degrade to single chart

Usage Example

Input:

{
  "data": [
    {"plan_type": "Premium Plan", "ticket_count": 47, "percentage": 33.6},
    {"plan_type": "Standard Plan", "ticket_count": 38, "percentage": 27.1},
    {"plan_type": "Basic Plan", "ticket_count": 25, "percentage": 17.9},
    {"plan_type": "Lite Plan", "ticket_count": 18, "percentage": 12.9},
    {"plan_type": "Enterprise Plan", "ticket_count": 12, "percentage": 8.6}
  ],
  "chart_type": "auto",
  "title": "Support Tickets by Plan - May 2026",
  "theme": "warm"
}

Execution flow: Phase 1 recommends -> bar (horizontal bar) -> Phase 2 generates ECharts config -> Phase 3 generates interactive HTML page

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.