Documentation Index
Fetch the complete documentation index at: https://docs.ai-stats.phaseo.app/llms.txt
Use this file to discover all available pages before exploring further.
Use this recipe when a Go service needs to answer operational questions with local context.
1. Install the SDKs
go get github.com/AI-Stats/AI-Stats/packages/sdk/sdk-go@latest
go get github.com/AI-Stats/AI-Stats/packages/sdk/agent-sdk-go@latest
statusTool := aistatsagent.DefineTool(aistatsagent.Tool{
ID: "read-status",
Description: "Read the current service status page summary.",
Execute: func(input any, ctx aistatsagent.RuntimeContext) (any, error) {
return map[string]any{
"status": "degraded",
"summary": "Latency is elevated on one provider route.",
}, nil
},
})
3. Create the agent
agent := aistatsagent.CreateAgent(aistatsagent.AgentDefinition{
ID: "ops-runbook-agent",
Model: "openai/gpt-5.4-nano",
Instructions: "Use tools when helpful and answer briefly.",
Tools: []aistatsagent.Tool{statusTool},
})
4. Run one bounded troubleshooting turn
client, _ := aistatsagent.CreateGatewayAgentClient(aistatsagent.GatewayAgentClientOptions{})
result, err := agent.Run(context.Background(), aistatsagent.RunOptions{
Input: "What is happening with provider latency right now?",
Client: client,
})
if err != nil {
panic(err)
}
fmt.Println(result.Output)
When this recipe fits
- the workflow belongs inside one Go service
- the supporting context already exists as local or internal calls
- you want an agent loop without introducing a larger job system yet
