​

AbstractAgent

The AbstractAgent class provides the foundation for all agent implementations in the Agent Wire Protocol. It handles the core event stream processing, state management, and message history.

import { AbstractAgent } from "@agentwire/client"

​

Configuration

By default, all agents are configured by providing an optional AgentConfig object to the constructor.

interface AgentConfig {
  agentId?: string // The identifier of the agent
  description?: string // A description of the agent, used by the LLM
  threadId?: string // The conversation thread identifier
  initialMessages?: Message[] // An array of initial messages
  initialState?: State // The initial state of the agent
}

​

Adding Configuration Options in your Subclass

To add additional configuration options, it is recommended to extend the AgentConfig interface and call the super constructor with the extended config from your subclass like this:

interface MyAgentConfig extends AgentConfig {
  myConfigOption: string
}

class MyAgent extends AbstractAgent {
  private myConfigOption: string

  constructor(config: MyAgentConfig) {
    super(config)
    this.myConfigOption = config.myConfigOption
  }
}

​

Core Methods

​

runAgent()

The primary method for executing an agent and receiving the event stream.

runAgent(parameters?: RunAgentParameters): Observable<RuntimeProtocolEvent>

​

Parameters

interface RunAgentParameters {
  runId?: string // Unique ID for this execution run
  tools?: Tool[] // Available tools for the agent
  context?: Context[] // Contextual information
  forwardedProps?: Record<string, any> // Additional properties to forward
}

​

abortRun()

Cancels the current agent execution.

abortRun(): void

​

clone()

Creates a deep copy of the agent instance.

clone(): AbstractAgent

​

Properties

• agentId: Unique identifier for the agent instance
• description: Human-readable description
• threadId: Conversation thread identifier
• messages: Array of conversation messages
• state: Current agent state object

​

Protected Methods

These methods are meant to be implemented or extended by subclasses:

​

run()

Executes the agent and returns an observable event stream.

protected abstract run(input: RunAgentInput): RunAgent

​

apply()

Processes events from the run and updates the agent state.

protected apply(input: RunAgentInput): ApplyEvents

​

prepareRunAgentInput()

Prepares the input parameters for the agent execution.

protected prepareRunAgentInput(parameters?: RunAgentParameters): RunAgentInput

​

onError() and onFinalize()

Lifecycle hooks for error handling and cleanup operations.

protected onError(error: Error): void
protected onFinalize(): void