Node Types

almyty agents support 9 node types. Each node performs a specific function within the pipeline and can be connected to other nodes via edges.

Input Node

The entry point of every pipeline. Defines the JSON Schema that incoming data must conform to.

Configuration:

{
  "id": "input_1",
  "type": "input",
  "data": {
    "schema": {
      "type": "object",
      "properties": {
        "message": { "type": "string" },
        "context": { "type": "string" }
      },
      "required": ["message"]
    }
  }
}

Every pipeline must have exactly one Input node. Access its data in templates via {{input.message}} or {{input.context}}.

Output Node

The exit point of the pipeline. Maps the final result from upstream nodes.

Configuration:

{
  "id": "output_1",
  "type": "output",
  "data": {
    "mapping": "{{nodes.llm_1.output}}"
  }
}

Every pipeline must have exactly one Output node.

LLM Call Node

Sends a prompt to a configured LLM provider and returns the response.

Configuration:

{
  "id": "llm_1",
  "type": "llm_call",
  "data": {
    "providerId": "provider-uuid",
    "model": "gpt-4o",
    "systemPrompt": "You are a helpful assistant.",
    "userPromptTemplate": "Answer this: {{input.message}}",
    "temperature": 0.5,
    "maxTokens": 2048,
    "toolIds": ["tool-1", "tool-2"]
  }
}

When toolIds are provided, the LLM can invoke those tools during execution. The tool call results are fed back to the LLM automatically.

Tool Call Node

Directly invokes a tool with explicit parameter mapping.

Configuration:

{
  "id": "tool_1",
  "type": "tool_call",
  "data": {
    "toolId": "tool-uuid",
    "parameterMapping": [
      { "key": "query", "value": "{{input.message}}" },
      { "key": "limit", "value": "10" }
    ]
  }
}

The result is available downstream as {{nodes.tool_1.output}}.

Condition Node

Branches the pipeline based on a boolean expression.

Configuration:

{
  "id": "cond_1",
  "type": "condition",
  "data": {
    "expression": "{{nodes.llm_1.output}}.includes('error')"
  }
}

A Condition node has two outgoing edges: one for true and one for false. Connect each branch to different downstream nodes.

Transform Node

Applies a JavaScript transformation to data flowing through the pipeline.

Configuration:

{
  "id": "transform_1",
  "type": "transform",
  "data": {
    "code": "const data = {{nodes.tool_1.output}};\nreturn data.map(item => item.name).join(', ');"
  }
}

Transform nodes run in a sandboxed environment. They receive the resolved template values and must return a value.

Loop Node

Iterates over an array, executing a sub-pipeline for each item.

Configuration:

{
  "id": "loop_1",
  "type": "loop",
  "data": {
    "iterableExpression": "{{nodes.tool_1.output.items}}",
    "maxIterations": 50
  }
}

Inside the loop body, access the current item as {{loop.item}} and the index as {{loop.index}}.

Parallel Node

Executes multiple downstream branches simultaneously and aggregates results.

Configuration:

{
  "id": "parallel_1",
  "type": "parallel",
  "data": {
    "branches": ["llm_1", "llm_2", "tool_1"],
    "aggregation": "array"
  }
}

Results are available as {{nodes.parallel_1.output}}, which is an array or object depending on the aggregation strategy.

Sub-Agent Node

Invokes another agent as a step within this pipeline.

Configuration:

{
  "id": "sub_1",
  "type": "sub_agent",
  "data": {
    "agentId": "agent-uuid",
    "inputMapping": [
      { "key": "query", "value": "{{nodes.llm_1.output}}" }
    ]
  }
}

The sub-agent's output is available as {{nodes.sub_1.output}}. This enables composable, modular agent architectures.