run_task

Execute advanced AI tasks using state-of-the-art models for reasoning, analysis, and more. Start single or batch tasks with real-time progress monitoring and retrieve results using task IDs. Ideal for complex workflows requiring AI-driven insights.

Instructions

Start a complex AI task. Perform advanced reasoning and analysis with state of the art LLMs. Start multiple tasks at once by using an array for model. Returns a task ID immediately (or batch ID for multiple models) to check status and retrieve results.

Input Schema

Name	Required	Description
`context`	No	Optional: Background context for the task
`files`	No	Optional: Array of file paths to include in the task context
`model`	No	Optional: Single model OR array of models for batch execution. Defaults to 'standard' if not specified.
`output`	No	Optional: The desired output/success state
`read_only`	No	Optional: When true, excludes tools that can modify files, execute commands, or make changes. Only allows read/search/analysis tools.
`task`	Yes	The task prompt - what to perform (required)

Input Schema (JSON Schema)

{
  "properties": {
    "context": {
      "description": "Optional: Background context for the task",
      "type": "string"
    },
    "files": {
      "description": "Optional: Array of file paths to include in the task context",
      "items": {
        "type": "string"
      },
      "type": "array"
    },
    "model": {
      "description": "Optional: Single model OR array of models for batch execution. Defaults to 'standard' if not specified.",
      "oneOf": [
        {
          "description": "Model class OR specific model name. Classes: reasoning, vision, standard, mini, reasoning_mini, code, writing, summary, vision_mini, long. Popular models: gpt-5, grok-4, gemini-2.5-pro, claude-opus-4-1.",
          "enum": [
            "reasoning",
            "vision",
            "standard",
            "mini",
            "reasoning_mini",
            "code",
            "writing",
            "summary",
            "vision_mini",
            "long",
            "gpt-5",
            "grok-4",
            "gemini-2.5-pro",
            "claude-opus-4-1"
          ],
          "type": "string"
        },
        {
          "description": "Array of model classes or specific model names for batch execution",
          "items": {
            "enum": [
              "reasoning",
              "vision",
              "standard",
              "mini",
              "reasoning_mini",
              "code",
              "writing",
              "summary",
              "vision_mini",
              "long",
              "gpt-5",
              "grok-4",
              "gemini-2.5-pro",
              "claude-opus-4-1"
            ],
            "type": "string"
          },
          "type": "array"
        }
      ]
    },
    "output": {
      "description": "Optional: The desired output/success state",
      "type": "string"
    },
    "read_only": {
      "default": false,
      "description": "Optional: When true, excludes tools that can modify files, execute commands, or make changes. Only allows read/search/analysis tools.",
      "type": "boolean"
    },
    "task": {
      "description": "The task prompt - what to perform (required)",
      "type": "string"
    }
  },
  "required": [
    "task"
  ],
  "type": "object"
}

MCP Task

run_task

Instructions

Input Schema

Input Schema (JSON Schema)

Other Tools from MCP Task

Related Tools

MCP directory API