KosmoKrator

rendering

Vega-Lite MCP Integration for OpenAI Agents SDK

Connect Vega-Lite to OpenAI Agents SDK through the local KosmoKrator MCP gateway with scoped tools, credentials, and write policy.

1 functions 0 read 1 write No credentials auth
Submit Vega-Lite feedback View package source

Connect Vega-Lite to OpenAI Agents SDK

Attach KosmoKrator integration tools to OpenAI Agents SDK workflows through a local MCP gateway.

Start the KosmoKrator MCP gateway locally and point the OpenAI Agents SDK MCP tool at that process or wrapper. The gateway is local, scoped to this integration, and starts with --write=deny so OpenAI Agents can inspect read-capable tools without receiving write access by default.

Vega-Lite MCP Config for OpenAI Agents SDK

Use headless JSON commands for CI-style execution and MCP for agent tool discovery.

{
  "mcpServers": {
    "kosmokrator-vegalite": {
      "type": "stdio",
      "command": "kosmo",
      "args": [
        "mcp:serve",
        "--integration=vegalite",
        "--write=deny"
      ]
    }
  }
}

Run the Gateway Manually

kosmokrator mcp:serve --integration=vegalite --write=deny

Why Use KosmoKrator Here

Scoped tools

Expose only Vega-Lite instead of a broad multi-service tool list.

Local credentials

Reuse credentials already configured for the KosmoKrator CLI and Lua runtime.

Write policy

Start read-only, then opt into ask or allow for trusted workspaces.

Vega-Lite Tools Visible to OpenAI Agents

OpenAI Agents sees stable MCP tool names generated from the Vega-Lite integration catalog.

MCP toolSource functionTypeDescription
integration__vegalite__render_vegalite vegalite.render_vegalite Write Render a Vega-Lite visualization to a PNG image. Pass a complete Vega-Lite JSON specification and get back a markdown image embed. IMPORTANT: Always use inline data with "data": {"values": [...]}. Never use "data": {"url": "..."}. Example spec: { "$schema": "https://vega.github.io/schema/vega-lite/v5.json", "data": {"values": [ {"category": "A", "value": 28}, {"category": "B", "value": 55}, {"category": "C", "value": 43} ]}, "mark": "bar", "encoding": { "x": {"field": "category", "type": "nominal"}, "y": {"field": "value", "type": "quantitative"} } } Supported mark types: bar, line, point, area, rect, circle, square, arc, text, tick, rule, trail, boxplot. Always include "type" in encoding channels: "quantitative", "nominal", "ordinal", or "temporal".

Related Vega-Lite Pages

Vega-Lite CLIUse direct JSON commands instead of MCP discovery. Vega-Lite MCPFull gateway reference and MCP tool names. KosmoKrator for OpenAI AgentsClient-level gateway setup.