Skip to main content
Workflow Studio is the visual editor for designing agent workflows. Open it from the Console sidebar at Workflow Studio (/workflow).

​
The authoring loop

1

Create a workflow

Click New workflow and give it a name. The canvas opens with an empty graph.
2

Add steps

Drag step types onto the canvas and connect them with dependencies. Each step needs a unique id, a type, and type-specific configuration.
3

Wire dependencies

Steps run when their depends_on predecessors complete. Independent branches run in parallel automatically.
4

Validate

Click Validate to run a server-side dry-run. Validation checks graph structure, dependency cycles, Lua syntax, and MCP bindings — without executing tools. Chained {{steps.*}} templates are checked for wiring and syntax only; they do not require upstream step outputs. See Dry-run semantics.
5

Import (optional)

Use Import to load a shared graph from JSON — canvas { "nodes", "edges" } or backend { "steps": [] }. The review step shows a read-only preview and, for MCP steps, lets you map each server_url to an instance in your workspace before applying to the canvas.
6

Edit as JSON (optional)

Switch the center panel to JSON mode to edit the backend graph with IntelliSense, Apply to canvas, and Edit script for Lua steps. See JSON editor.
7

Publish

Publish an immutable version when the graph is ready. Runs always execute a published snapshot, not draft edits.
8

Run

Start a run from the workflow configuration panel (Run setup). Configure shared input (trigger_payload) and optional per-step overrides (trigger_steps), then Run or Dry-run. Use Save defaults to persist start values on the workflow version (separate from canvas Save).

​
Template variables

Steps pass data downstream through template strings:
Use these in tool_args, LLM prompt fields, for_each_items, and Lua script inputs. Upstream step results are available under steps.<step-id>.result. Studio suggests variables as you type — see Template intellisense for search tips, the variable picker, and the Variables reference tab. Run-start input uses {{input.*}} (from trigger_payload and optional trigger_steps). See Run setup and start input.

​
Versioning

Each publish creates a new immutable version. You can list versions, roll back runs to a specific version, and compare graphs. Draft edits do not affect in-flight runs.

​
Live streaming

While a run is active, Workflow Studio subscribes to a WebSocket event stream (/v1/workflows/{id}/stream). You see step start, progress, completion, failure, and human-task pause events as they happen.

​
Autopilot assistance

The Console Chat panel (Autopilot) can scaffold starter workflows, create graphs from natural language, and navigate you to Workflow Studio. Autopilot actions include create_workflow, scaffold_starter, and run_workflow.

​
Archive workflows

From the workflow list, select a workflow and click Archive to remove it from your active Studio list. Archive is soft — the definition is retained for run history and audit, not permanently erased.
A dedicated Archived view and Restore action are planned but not available yet. True permanent delete will be a separate action from archive in a future release.

​
Next steps