Skip to main content

​
1. Create an account

Sign up at app.flowyble.com and create your organization. Every new organization starts on the Free plan with a generous credit allowance.

​
2. Create your first agent

1

Go to Agents

Navigate to Agents in the sidebar and click Create Agent.
2

Configure the agent

  • Name: Give your agent a descriptive name (e.g. “Email Summarizer”)
  • LLM Model: Choose a model (e.g. GPT-4o, Claude Sonnet)
  • Behavior Prompt: Describe what the agent should do — this is the system prompt that guides its reasoning
  • Max Steps: Set the maximum number of tool-call iterations (1–12, default 8)
3

Attach tools (optional)

Add built-in functions or custom tools that the agent can call during execution. For example, attach a SearchKnowledge tool to let the agent query a knowledge base.
4

Save and activate

Set the agent status to Active and save.

​
3. Run the agent

​
From the dashboard

Go to the agent detail page and click Run. Enter a user message and watch the execution in real time.

​
Via the API

curl -X POST https://api.flowyble.com/v1/agents/run/{agentId} \
  -H "X-API-Key: your-api-key" \
  -H "Content-Type: application/json" \
  -d '{"userMessage": "Summarize my latest emails"}'
The response contains a runId you can use to poll for status: 
curl https://api.flowyble.com/v1/agents/run/{runId}/status \
  -H "X-API-Key: your-api-key"

​
4. View results

Each agent run produces:
  • Final Output — the agent’s answer or result
  • Steps — a step-by-step log of tool calls, reasoning, inputs, and outputs
  • Metrics — tokens used, credits consumed, execution duration

​
Next steps

Core Concepts

Understand agents, tools, runs, and the credit system.

API Reference

Integrate Flowyble into your applications.