Python SDK for AI agent monitoring, LLM cost tracking, benchmarking, and more. Integrates with most LLMs and agent frameworks including CrewAI, Agno, OpenAI Agents SDK, Langchain, Autogen, AG2, and CamelAI
Python agentagentopsagents-sdkaianthropicautogencost-estimationcrewaievalsevaluation-metricsgroqlangchainllmmistralollamaopenaiopenai-agents
export type ActivityType = 'LLM' | 'Action' | 'Tool' | 'Error';
export interface TimelineActivity {
start: number; // Start time in seconds
end: number; // End time in seconds
type: ActivityType; // Type of activity
label?: string; // Optional label to show on activity
details?: string; // Optional additional details for tooltip
id?: string; // ID of the activity (agent_id)
rawData?: any; // Original raw data object
}
export interface TimelineLane {
id: string; // Unique identifier for the lane
name: string; // Display name of the lane
color?: string; // Optional background color for the lane
activities: TimelineActivity[]; // Activities in this lane
}
export interface TimelineData {
lanes: TimelineLane[];
maxTime?: number; // Optional maximum time in seconds for the chart
}
// New types for the provided data format
export interface AgentActivityData {
agent_id: string;
type: string;
Agent: string;
Start: number;
End: number;
Duration: number;
Name?: string;
Model?: string;
Cost?: string;
'Prompt Tokens'?: number;
'Completion Tokens'?: number;
'Promptarmor Flag'?: string | null;
Thread?: string | null;
Prompt?: any;
Completion?: any;
Params?: any;
Returns?: any;
}
export interface AgentData {
id: string;
data: AgentActivityData[];
}
export type AgentsTimelineData = AgentData[];