How Do I Set Up AI Content Drafting with Client Approval at My Agency?

The steps

1. 1

   Create the content-drafting agent

   An agent in Dave by voolama (hellodave.ai) is a reusable AI component that wraps a provider with a system prompt, a model choice, and behavior parameters. Every Agent Interaction node in a workflow references a named agent and a specific version. Create the agent first so it is available to select when you build the workflow graph. Sourced from the agents help documentation at adminrodeo.qubiccontent.com/dashboard/help/agents (DEV v0.11.0 R-001, 2026-08-30).

   Only Admin and Create roles can create agents and agent versions. Dave by voolama has six default roles: Admin, Create, Curate, Use, Reporting, and Financial. Roles are editable.

   Do this

   • Navigate to Agents in the sidebar. Click New Agent.
   • Enter a descriptive name: for example, "Social Caption Drafter" or "Blog Intro Writer". Agent names should describe the function, not the underlying model.
   • Optionally add a Description and Tags (comma-separated, for example: content, agency, social).
   • Click Create Agent. The agent appears in the list with zero versions.
   • Click the agent name to open its detail page. Click Create Version.
   • Enter a valid JSON object in the Config (JSON) editor. Example for a social caption agent: {"systemPrompt": "You are a social media copywriter for a small business. Write one Instagram caption of 60 words or fewer based on the brief in the workflow context. Use a friendly, direct tone. Do not use hashtags.", "model": "gpt-4o", "temperature": 0.7, "maxTokens": 200}
   • Click Create Version. The version appears as v1. The version number is auto-assigned and cannot be changed.

   Example

   A hypothetical agency creates an agent named "Caption Drafter" with a system prompt instructing the AI to write one Instagram caption of 60 words or fewer, friendly tone, no hashtags. She sets model to gpt-4o and temperature to 0.7. She validates the JSON in a browser console using JSON.parse() before submitting. The agent is now at v1 and available to select in the Workflow Editor. See the Create an Agent quick-reference for a visual walkthrough.

   Best practice

   The agents help documentation states: 'Version before editing. Always create a new version rather than mentally tracking changes. Old versions remain available if you need to roll back, and the immutable version history gives you a clear audit trail.' Name agents by function, not by model: "Caption Drafter" survives a model change; "GPT-4o Agent" does not. Keep Config JSON minimal: only include fields your provider actually uses. Retire old versions once a new version is confirmed working, to prevent team members from accidentally selecting the old one in new workflows.

2. 2

   Build the five-node workflow graph

   The Workflow Editor in Dave by voolama (hellodave.ai) is a visual drag-and-drop canvas. No code is required. Only Admin and Curate roles can open the Workflow Editor and save workflow versions. Sourced from the workflow-editor help documentation at adminrodeo.qubiccontent.com/dashboard/help/workflow-editor (DEV v0.11.0 R-001, 2026-08-30).

   Node	Type	Label to use	Purpose	Creates a task?
   1	Start	Start	Entry point. Every workflow must have exactly one Start node.	No
   2	User Interaction	Collect Brief	Account manager submits the campaign brief. The workflow pauses and creates a task in the Task Inbox.	Yes
   3	Agent Interaction	Draft Content	AI agent runs automatically using the brief from the previous task. No human involvement at this node.	No
   4	Human Review	Account Manager Review	Account manager reads the AI draft and approves, rejects, or requests changes before it goes to the client.	Yes
   5	Human Review	Client Approval	Client reads the approved draft and gives final sign-off before delivery.	Yes
   6	End	Delivered	Terminal node. Marks workflow completion after client approval.	No

   Do this

   • Navigate to Workflows in the sidebar. Click Create Workflow. Name it (for example, "Caption Approval Pipeline") and click Save.
   • Click into the workflow to open the Workflow Editor.
   • From the Node Palette on the left, drag a Start node onto the canvas.
   • Drag a User Interaction node below Start. In the Properties Panel on the right, rename it "Collect Brief".
   • Drag an Agent Interaction node below. Rename it "Draft Content". In the Properties Panel, select the agent you created in Step 1 (for example, "Caption Drafter") and its v1 version.
   • Drag a Human Review node below. Rename it "Account Manager Review".
   • Drag a second Human Review node below. Rename it "Client Approval".
   • Drag an End node below. Rename it "Delivered".
   • Connect the nodes: hover over the source handle (bottom circle) of Start, drag to the target handle (top circle) of Collect Brief. Repeat for each pair in sequence down to Delivered.
   • Click Save Version. The workflow is now at v1. The node/edge counter in the top-right shows 6 nodes, 5 edges.

   Example

   A hypothetical agency owner drags six nodes onto the canvas, connects them in sequence, and clicks Save Version. The counter shows 6 nodes, 5 edges. She clicks the Agent Interaction node and confirms the Properties Panel shows "Caption Drafter v1" in the agent assignment section. See the Build Your First Workflow quick-reference for a visual walkthrough of the Workflow Editor.

   Best practice

   The workflow-editor help documentation states: 'Always start with a Start node. Every workflow requires exactly one Start node as its entry point.' It also states: 'Ensure all paths lead to an End node. Dangling paths may cause instances to hang at runtime.' Name nodes descriptively: "Account Manager Review" is self-documenting; "Human Review" is not. Save frequently: each save creates a new immutable version, so there is no risk of overwriting your work. The editor does not auto-save: always click Save Version before navigating away.

3. 3

   Configure the Request Changes loop

   By default, each Human Review node has one outgoing edge (the Approve path). To handle Reject and Request Changes decisions, you need to add additional outgoing edges from each Human Review node and connect them to the appropriate next step. The Workflow Editor supports multiple outgoing edges from a single node. Sourced from the workflow-editor help documentation, DEV v0.11.0, confirmed 2026-08-30.

   The tasks help documentation states: 'Your decision (Approve, Reject, Request Changes) determines which outgoing edge the workflow follows from the review node. This means your choice directly affects the path the instance takes next.'

   Do this

   • Open the Workflow Editor for the workflow you saved in Step 2.
   • Click the Account Manager Review node to select it.
   • Hover over the source handle (bottom circle) of the Account Manager Review node. Drag a new edge back to the Draft Content (Agent Interaction) node. This creates the Request Changes loop: when the account manager requests changes, the AI agent runs again.
   • Optionally add a Routing node between the Account Manager Review node and the Draft Content node to label the path and add routing criteria. The Routing node is a logic-only node: it evaluates criteria and routes execution without creating tasks or invoking agents.
   • Repeat for the Client Approval node: drag a new edge from Client Approval back to Account Manager Review for the Request Changes path, so the account manager sees the revised draft before it goes back to the client.
   • Click Save Version. The workflow is now at v2 with the loop edges included.

   Example

   A hypothetical agency owner adds a Request Changes edge from Account Manager Review back to Draft Content. She adds a second Request Changes edge from Client Approval back to Account Manager Review. She clicks Save Version. The counter now shows 6 nodes, 7 edges. She verifies: if the client requests changes, the workflow loops back to the account manager for a second internal review before the AI re-drafts.

   Best practice

   The workflow-editor help documentation states: 'Use Information nodes for documentation. Drop Information nodes next to complex sections of your graph to explain the intent.' Add an Information node next to the Request Changes loop with a note explaining the loop logic. This makes the graph self-documenting for future editors. For a simple first workflow, the Approve path alone is sufficient: add the loop edges once the basic workflow is running correctly.

4. 4

   Set the workflow to active and launch the first instance

   A workflow must be set to active status before team members can launch instances. A draft workflow cannot be run. Only Admin and Curate roles can change workflow status from draft to active. Sourced from the workflow-editor help documentation, DEV v0.11.0, confirmed 2026-08-30.

   Do this

   • Navigate to Workflows in the sidebar. Find the workflow you built.
   • Change the workflow status from draft (gray badge) to active (green badge). Only Admin and Curate roles can do this.
   • Navigate to Instances in the sidebar. Click Create Instance.
   • Select the workflow (for example, "Caption Approval Pipeline") and its latest version. Click Launch.
   • Navigate to Tasks. Open the Collect Brief task. Read the Payload section to understand what input the workflow is requesting. Type the campaign brief and click Approve to submit it.
   • The Agent Interaction node runs automatically. Refresh Tasks after a few seconds.
   • Open the Account Manager Review task. Read the Payload: it contains the AI-drafted content. Click Approve if the draft is good, Request Changes if it needs revision, or Reject if it is wrong. Add a Comment explaining your decision.
   • If you approved, the Client Approval task appears next. The client opens it, reads the draft, and clicks Approve to complete the workflow. The instance status changes to completed.

   Example

   A hypothetical account manager launches the Caption Approval Pipeline, submits a brief for a summer menu promotion, and waits. Thirty seconds later an Account Manager Review task appears. The AI has drafted a 55-word caption. The account manager reads it, adds a comment noting one word change, and clicks Approve. A Client Approval task appears. The client reads the revised caption and clicks Approve. The instance status changes to completed. See the Launch an Instance and Work a Task from Your Inbox quick-references.

   Best practice

   The tasks help documentation states: 'Use comments to document your decisions. Even a brief note explaining why you rejected or requested changes helps workflow owners and other reviewers understand the history.' For an agency, the comment is the reviewer's professional judgment on record. Every comment is stored in the Submitted Result field of the completed task and is visible in the audit log export. For content that includes client endorsements or testimonials, 16 CFR Part 255.5 (ecfr.gov/current/title-16/chapter-I/subchapter-B/part-255) requires disclosure of material connections: the audit record of the Human Review decision is the agency's evidence that a named person reviewed and approved the content, but it is not a substitute for the required disclosure in the content itself.

Roles and who does what in a content approval workflow in Dave by voolama

What this guide does not cover