Skip to main content
Glama
VisActor

vchart-mcp-server

Official
by VisActor

generate_sankey_chart

Visualize flow relationships between nodes in complex networks to display distribution and flow paths of source and target data using Sankey diagrams.

Instructions

Generate a Sankey diagram to visualize the flow relationships between nodes in complex networks, suitable for displaying the distribution and flow paths of source and target data.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
outputNoChart output type. Defaults to 'image'.image
widthNoChart width. Optional, defaults to 500.
heightNoChart height. Optional, defaults to 500.
dataTableYesData for the Sankey diagram, e.g., [{ category: 'category 01', value: 10 }].
sourceFieldYesThe source field in the Sankey diagram; must exist in the data.
targetFieldYesThe target field in the Sankey diagram; must exist in the data.
valueFieldYesMeasure field. Must be numeric and exist in the data.
chartThemeNoChart theme. Optional, defaults to 'light'.
titleNoChart title text.
subTitleNoChart subtitle text.
titleOrientNoTitle position in the chart.
backgroundNoChart background color (hex). Optional, defaults to white.
colorsNoColor palette for chart elements.

Implementation Reference

  • The central MCP CallToolRequestHandler that handles execution for all chart tools, including 'generate_sankey_chart'. It maps tool name to chartType ('sankey'), validates input with the specific Zod schema, calls generateChartByType('sankey', args), and returns the chart output (spec, image URL, or HTML).
    server.setRequestHandler(CallToolRequestSchema, async request => {
      const toolName = request.params.name;
      const chartType = Object.keys(Charts).find(
        key => (Charts as any)[key].tool.name === toolName
      );
    
      if (!chartType) {
        throw new McpError(
          ErrorCode.MethodNotFound,
          `Unknown tool: ${request.params.name}.`
        );
      }
    
      try {
        // Validate input using Zod before generating chart
        const args = request.params.arguments || {};
    
        // Select the appropriate schema based on the chart type
        const schema = Charts[chartType as keyof typeof Charts].schema;
    
        if (schema) {
          const result = schema.safeParse(args);
          if (!result.success) {
            throw new McpError(
              ErrorCode.InvalidParams,
              `Invalid parameters: ${result.error.message}`
            );
          }
        }
    
        const res = await generateChartByType(chartType, args);
    
        if (res && (res as any).spec) {
          return {
            content: [
              {
                type: 'text',
                text: JSON.stringify((res as any).spec, null, 2),
              },
            ],
          };
        }
    
        if (res && (res as any).image) {
          return {
            content: [
              {
                type: 'text',
                text: (res as any).image,
              },
            ],
          };
        }
    
        if (res && (res as any).html) {
          return {
            content: [
              {
                type: 'text',
                text: (res as any).html,
              },
            ],
          };
        }
    
        return {
          content: [
            {
              type: 'text',
              text: 'Failed to generate chart',
            },
          ],
        };
      } catch (error: any) {
        if (error instanceof McpError) {
          throw error;
        }
        throw new McpError(
          ErrorCode.InternalError,
          `Failed to generate chart: ${error?.message || 'Unknown error.'}`
        );
      }
    });
  • Core helper function generateChartByType that implements the chart generation logic for all types including Sankey. Processes input options (sourceField -> cell.source, etc.), generates VChart spec using @visactor/generate-vchart, and renders to image/HTML/spec via a remote server.
    export async function generateChartByType(chartType: string, options: any) {
      const {
        title,
        subTitle,
        titleOrient,
        xAxisType,
        xAxisOrient,
        xAxisTitle,
        xAxisHasGrid,
        xAxisHasLabel,
        xAxisHasTick,
    
        yAxisType,
        yAxisOrient,
        yAxisTitle,
        yAxisHasGrid,
        yAxisHasLabel,
        yAxisHasTick,
    
        leftYAxisTitle,
        leftYAxisHasGrid,
        leftYAxisHasLabel,
        leftYAxisHasTick,
    
        rightYAxisTitle,
        rightYAxisHasGrid,
        rightYAxisHasLabel,
        rightYAxisHasTick,
    
        angleAxisTitle,
        angleAxisHasGrid,
        angleAxisHasLabel,
        angleAxisHasTick,
        angleAxisType,
    
        radiusAxisHasGrid,
        radiusAxisHasLabel,
        radiusAxisHasTick,
        radiusAxisType,
        radiusAxisTitle,
    
        output,
        width,
        height,
        ...res
      } = options;
    
      const opts = { ...res };
      const titleObj = filterValidAttributes({
        text: title,
        subText: subTitle,
        orient: titleOrient,
      });
      const xAxisObj = filterValidAttributes({
        type: xAxisType,
        orient: xAxisOrient,
        title: xAxisTitle,
        hasGrid: xAxisHasGrid,
        hasLabel: xAxisHasLabel,
        hasTick: xAxisHasTick,
      });
      const yAxisObj = filterValidAttributes({
        type: yAxisType,
        orient: yAxisOrient,
        title: yAxisTitle,
        hasGrid: yAxisHasGrid,
        hasLabel: yAxisHasLabel,
        hasTick: yAxisHasTick,
      });
      const leftYAxisObj = filterValidAttributes({
        title: leftYAxisTitle,
        hasGrid: leftYAxisHasGrid,
        hasLabel: leftYAxisHasLabel,
        hasTick: leftYAxisHasTick,
      });
      const rightYAxisObj = filterValidAttributes({
        title: rightYAxisTitle,
        hasGrid: rightYAxisHasGrid,
        hasLabel: rightYAxisHasLabel,
        hasTick: rightYAxisHasTick,
      });
      const angleAxisObj = filterValidAttributes({
        title: angleAxisTitle,
        hasGrid: angleAxisHasGrid,
        hasLabel: angleAxisHasLabel,
        hasTick: angleAxisHasTick,
        type: angleAxisType,
      });
      const radiusAxisObj = filterValidAttributes({
        hasGrid: radiusAxisHasGrid,
        hasLabel: radiusAxisHasLabel,
        hasTick: radiusAxisHasTick,
        type: radiusAxisType,
        title: radiusAxisTitle,
      });
    
      const cell: Record<string, string> = {};
    
      [
        "xField",
        "yField",
        "colorField",
        "categoryField",
        "valueField",
        "wordField",
        "sizeField",
        "timeField",
        "sourceField",
        "targetField",
        "setsField",
        "radiusField",
      ].forEach((fieldName) => {
        if (isValid(options[fieldName])) {
          cell[fieldName.replace("Field", "")] = options[fieldName];
          delete opts[fieldName];
        }
      });
    
      opts.cell = cell;
    
      if (!isEmpty(titleObj)) {
        opts.title = titleObj;
      }
    
      const axes = [
        xAxisObj,
        yAxisObj,
        leftYAxisObj,
        rightYAxisObj,
        angleAxisObj,
        radiusAxisObj,
      ];
    
      if (axes.some((axis) => !isEmpty(axis))) {
        opts.axes = axes.filter((axis) => !isEmpty(axis));
      }
    
      const { spec } = generateChart(options.chartType ?? chartType, opts);
    
      if (!spec) {
        return null;
      }
    
      if (output === "spec") {
        if (isValid(width)) {
          spec.width = width;
        }
        if (isValid(height)) {
          spec.height = height;
        }
    
        return {
          spec: spec,
        };
      }
    
      return gentrateChartImageOrHtml(output, spec, {
        width: `${width ?? 500}px`,
        height: `${height ?? 500}px`,
      });
    }
  • Zod schema defining the input parameters and validation for the generate_sankey_chart tool.
    const schema = z.object({
      output: ChartOutputSchema,
      width: WidthSchema,
      height: HeightSchema,
      dataTable: z
        .array(z.any())
        .describe(
          "Data for the Sankey diagram, e.g., [{ category: 'category 01', value: 10 }]."
        )
        .nonempty({ message: "Sankey diagram data cannot be empty." }),
    
      sourceField: z
        .string()
        .nonempty({ message: "Source field cannot be empty." })
        .describe(
          "The source field in the Sankey diagram; must exist in the data."
        ),
      targetField: z
        .string()
        .nonempty({ message: "Target field cannot be empty." })
        .describe(
          "The target field in the Sankey diagram; must exist in the data."
        ),
      valueField: YFieldSchema,
    
      chartTheme: ThemeSchema,
      title: TitleTextSchema,
      subTitle: TitleSubTextSchema,
      titleOrient: TitleOrientSchema,
    
      background: BackgroundSchema,
      colors: ColorsSchema,
    });
  • src/server.ts:34-38 (registration)
    MCP ListToolsRequestHandler that registers and lists all chart tools, including generate_sankey_chart, by exporting tool metadata from Charts module.
    server.setRequestHandler(ListToolsRequestSchema, async () => {
      return {
        tools: Object.values(Charts).map(chart => (chart as any).tool),
      };
    });
  • Tool metadata definition and export for dynamic registration in the MCP server.
    const tool = {
      name: "generate_sankey_chart",
      description:
        "Generate a Sankey diagram to visualize the flow relationships between nodes in complex networks, suitable for displaying the distribution and flow paths of source and target data.",
      inputSchema: convertZodToJsonSchema(schema),
    };
    
    export const sankey = {
      schema,
      tool,
    };
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden of behavioral disclosure. It states the tool generates a diagram but lacks critical details: it doesn't specify if this is a read-only or mutating operation, what the output looks like (e.g., file format, size), performance considerations, or error handling. For a tool with 13 parameters and no annotations, this is a significant gap in transparency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, well-structured sentence that efficiently conveys the core purpose. It avoids redundancy and is front-loaded with the main action. However, it could be slightly more concise by omitting 'suitable for displaying...' which is somewhat repetitive, but overall, it's appropriately sized with zero wasted words.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (13 parameters, no output schema, no annotations), the description is incomplete. It fails to address key contextual aspects: what the output entails (e.g., a file, URL, or raw data), how errors are handled, or any behavioral traits like rate limits. For a chart generation tool with many options, more guidance is needed to ensure the agent can use it effectively.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, meaning all parameters are documented in the schema itself. The description adds no specific parameter information beyond implying the tool handles 'source and target data,' which loosely relates to sourceField and targetField. Since the schema does the heavy lifting, the baseline score of 3 is appropriate, as the description provides minimal additional semantic value.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'Generate a Sankey diagram to visualize the flow relationships between nodes in complex networks.' It specifies the verb ('generate'), resource ('Sankey diagram'), and context ('visualize flow relationships'), making it easy to understand. However, it doesn't explicitly differentiate from sibling tools like generate_hierarchical_chart or generate_heatmap_chart, which might also handle network or flow data, preventing a perfect score.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives. It mentions the tool is 'suitable for displaying the distribution and flow paths of source and target data,' but this is a restatement of purpose rather than usage advice. There's no mention of prerequisites, scenarios where other chart types might be better, or limitations, leaving the agent without operational context.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/VisActor/vchart-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server