AI CallTool

AI CallTool
Certified
Call a tool on an MCP server and return its result

Connects to a Model Context Protocol (MCP) server, invokes one of its tools with the given arguments, and returns the result — no LLM involved. Use io.kestra.plugin.ai.mcp.ListTools to discover the available tools and their argument schemas first.

yaml
type: io.kestra.plugin.ai.mcp.CallTool

Examples

Call the add tool on an MCP server over Streamable HTTP.

yaml
id: mcp_call_tool
namespace: company.ai

tasks:
  - id: call
    type: io.kestra.plugin.ai.mcp.CallTool
    url: https://mcp.example.com/mcp
    headers:
      Authorization: "Bearer {{ secret('MCP_API_TOKEN') }}"
    tool: add
    arguments:
      a: 5
      b: 12

Properties

tool *string

Name of the tool to call

url *string

URL of the MCP server

The Streamable HTTP or SSE endpoint of the MCP server.

arguments object

Arguments passed to the tool

failOnToolError booleanstring

Defaulttrue

Fail on tool error

Whether the task should fail when the MCP server reports the tool call as an error. When false, the error is instead reported in errorMessage and isError.

headers object

Custom headers

Useful, for example, for adding authentication tokens via the Authorization header.

logRequests booleanstring

Defaultfalse

Log requests

logResponses booleanstring

Defaultfalse

Log responses

pluginDefaultsRef string

Reference (ref) of the pluginDefaults to apply to this task.

timeout string

Connection timeout duration

When not set, the underlying MCP client's default timeout applies (no timeout is enforced by this task).

transport string

DefaultSTREAMABLE_HTTP

Possible Values

STREAMABLE_HTTPSSEUNKNOWN

Transport used to connect to the MCP server

Outputs

errorMessage string

Error message

Populated only when failOnToolError is false and the call failed.

isError boolean

Whether the tool call returned an error

result string

Tool result

The tool's text result.

structuredContent object

Structured content

The tool's structured content, when the server returned one.

Metrics

mcp.tool.calls counter

Unitcalls

Number of MCP tool calls, tagged by tool name

Mistral

Tasks and triggers for Mistral AI: chat completions and workflow orchestration.

DeepSeek

Tasks that call DeepSeek chat models for conversational or schema-guided JSON responses.

AI CallToolCertifiedCall a tool on an MCP server and return its result

More Plugins in this Category

Mistral

DeepSeek

AI CallTool
Certified
Call a tool on an MCP server and return its result