Appearance
What is Dino?
Dino helps you design, publish, and operate AI voice agents through one guarded workflow.
The main journey
The model manages the conversation. Dino manages the flow, tools, calls, and results.
text
[1] Design ──▶ [2] Validate ──▶ [3] Publish ──▶ [4] Test
│
▼
[7] Results ◀── [6] Call ◀── [5] Process- Design one business outcome and its decision points.
- Validate the complete agent source.
- Publish a new agent or evolve an existing agent.
- Test the exact published version.
- Choose Batch, Drip, or Inbound for repeatable calls.
- Let Dino run each call with the selected infrastructure.
- Read structured results or receive Drip callbacks.
Two ways to work
The AI-guided route is usually faster. The CLI route remains useful for direct control and automation.
Prompt your AI agent
Recommendedtext
Use the installed `dino-ping` and `dino-pong` skills to onboard me.
Check the selected workspace, my account role, and the available capabilities.
Do not change anything yet.
Explain which agent and call workflows my role can use.
Return the safest next step for my use case.Use the Dino CLI directly
text
dino version check
dino capabilities
dino workspace list
dino workspace current
dino workspace check
dino account getThe important boundaries
- An agent defines conversation behavior, tools, audio, and model settings.
- A process defines how repeatable calls use an agent and infrastructure.
- A published agent does not start a campaign.
- A process does not change the agent's flow.
- Every state change requires approval for the exact action.
- A preview or queued contact can place a real, billable call.
Start with an example
Review the agent examples before you design a new agent. Open the selected source tree through the installed dino-ping skill. The examples cover routing, dispatch, post-call capture, memory, and integrations.
Next
Use the quickstart for the complete first-agent path.