Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.cogniagent.ai/llms.txt

Use this file to discover all available pages before exploring further.

When you click an actor on the canvas, the config panel opens on the right. This is where you tell the agent who it is, what it knows, and what it should do.
Actor config panel

​
Name and icon

At the top of the panel:
  • Icon — Click the icon to pick from a Lucide icon library. The platform also auto-suggests an icon based on the actor’s name and context. You can change it any time.
  • Name — Short and descriptive. Appears on the canvas and in conversation transcripts as the “speaker”. Good: Pricing Specialist, Returns Intake, Tier 2 Support. Bad: Agent A, Bot.

​
Basic tab — the four main fields

​
Context

Background knowledge the actor needs. Inheritable by children. See Context and inheritance for the full story. 
You are the product expert for Acme Inc. You know everything about
Acme Pro and Acme Enterprise — features, integrations, security,
supported channels, roadmap.
Don’t repeat content that lives in Global Context. The agent will inherit it automatically if Global flow context is checked under Inherit Context.

​
Inherit Context

Checkboxes that let this actor opt into upstream context:
  • Global flow context — the globalContext from Flow settings.
  • <Each ancestor actor> — any actor that routes into this one.
  • Collected information so far — slot summary (default: on). See slots.

​
Instructions

What this specific actor should do. Private — never inherited by children. 
Answer product questions in clear, plain language. If the user asks about
pricing or plans, hand off — don't quote prices.
This is where you put the actor’s persona, its task, and any handoff rules.

​
Definition of Done

A short rule for when the actor’s job is complete. 
The user has a clear answer about pricing or has been booked for an
Enterprise quote call.
When the actor decides it has satisfied this rule, it signals completion. The platform then either hands off or closes the conversation.
Strict-mode actors must have a Definition of Done. Without one, the actor won’t be allowed to hand off — and the conversation will stall. See Focus modes.
If you leave it blank, the actor decides on its own when it’s finished. That’s usually fine for short, single-purpose actors. Set it explicitly for anything multi-step.

​
Focus Mode

Auto / Flexible / Persistent / Strict. Controls how strictly the actor sticks to its job when users change topics. Default is Auto. See Focus modes for the deep dive.

​
Capabilities

Click Configure Capabilities to open the capabilities drawer, where you wire up:
  • Knowledge bases — give the actor search access to your documents.
  • Integrations — let the actor send Slack messages, send Gmail emails, look up Shopify orders, etc.
  • Custom tools — run Python, call a workflow, hit an HTTP endpoint.
See Capabilities for the full guide.

​
Advanced tab — the step builder

The Basic tab gives you a free-form actor — one that thinks and replies turn by turn, deciding what to say based on its prompt. The Advanced tab adds a step builder below the basic fields. There you can scripted the actor’s behaviour step by step: 
flow-start
   ↓
ask-question  → "What's your name?"
   ↓
ask-question  → "What's your email?"
   ↓
ai-act        → "Send a confirmation email using the send-email tool"
   ↓
end
When steps are present, the actor runs scripted. When they’re absent, the actor runs free-form. See Step builder for when to use each.

​
Save changes

Click Save Changes at the bottom of the panel. Changes are picked up on the actor’s next turn — you can edit a deployed actor’s prompt mid-conversation and the next reply will reflect the change.
If you’re iterating on a prompt, keep a Test session open in another tab. Edit → save → send the next message — you’ll see the new behaviour in seconds.

​
Delete an actor

Click the actor, then either:
  • Use the trash icon that appears when the actor is selected.
  • Press Delete on your keyboard.
Deleting an actor also deletes every edge connected to it.

​
Anti-patterns

One actor that does five things. If your Instructions block is more than a couple of paragraphs, split into multiple actors with routing edges between them. Focused actors route better and produce better replies.
Instructions in Context. Behaviour rules (“always confirm before sending”) belong in Instructions. If you put them in Context, child actors will inherit and start trying to confirm too. Almost never what you want.

​
Next

Routing edges

Connect actors and define when handoffs happen.

Capabilities

Give the actor knowledge, integrations, and custom tools.

Step builder

Script an actor’s behaviour step by step.

Flow settings

Configure flow-wide context, knowledge, and integrations.