@agenticforge/workflow

Standalone DAG workflow engine. Powers WorkflowAgent and can be used directly without an agent wrapper.

Installation

npm install @agenticforge/workflow

What's included

• WorkflowEngine — DAG executor with topological sort, parallel waves, branch and loop support
• All workflow types: WorkflowDefinition, WorkflowNode, WorkflowResult, NodeResult, etc.

Node types

Type	Description
tool	Call a registered tool; inputTemplate supports {var} interpolation
llm	Call the LLM; promptTemplate supports {var} interpolation
fn	Custom async function (ctx, llm, registry) => string
passthrough	Forward a context value unchanged
branch	Conditional branch: condition(ctx) returns branch name
loop	do-while loop over body sub-DAG

Direct usage

import { WorkflowEngine } from "@agenticforge/workflow";
import { LLMClient } from "@agenticforge/core";

const engine = new WorkflowEngine({
  llm: new LLMClient({ provider: "openai", model: "gpt-4o" }),
  verbose: true,
  maxConcurrency: 4,
});

const result = await engine.execute(
  {
    name: "bilingual-report",
    nodes: [
      { id: "fetch",     type: "tool", toolName: "search", inputTemplate: "{input}",                      depends: [] },
      { id: "analyze",   type: "llm",  promptTemplate: "Analyze:\n{fetch}",                               depends: ["fetch"] },
      { id: "translate", type: "llm",  promptTemplate: "Translate to Chinese:\n{fetch}",                  depends: ["fetch"] },
      { id: "report",    type: "llm",  promptTemplate: "Bilingual report:\n{analyze}\n\n{translate}",      depends: ["analyze", "translate"] },
    ],
  },
  "State of AI in 2026",
);

console.log(result.output);
console.log(result.nodeResults); // per-node timing and status

WorkflowEngineOptions

Option	Type	Default	Description
llm	LLMClient	required	LLM instance
registry	ToolRegistry	—	Required for tool nodes
verbose	boolean	false	Log execution waves
maxConcurrency	number	unlimited	Max concurrent nodes per wave

Using with WorkflowAgent

For agent-wrapped usage with conversation history and hooks:

import { WorkflowAgent } from "@agenticforge/agents";
import type { WorkflowDefinition } from "@agenticforge/workflow";

const agent = new WorkflowAgent({ name: "my-workflow", llm, verbose: true });
agent.setWorkflow(definition);
const output = await agent.run("input text");

See the Agents Guide for full WorkflowAgent documentation.