Class ZeroShotAgent

Agent for the MRKL chain.

Example


const agent = new ZeroShotAgent({
  llmChain: new LLMChain({
    llm: new ChatOpenAI({ temperature: 0 }),
    prompt: ZeroShotAgent.createPrompt([new SerpAPI(), new Calculator()], {
      prefix: `Answer the following questions as best you can, but speaking as a pirate might speak. You have access to the following tools:`,
      suffix: `Begin! Remember to speak as a pirate when giving your final answer. Use lots of "Args"
Question: {input}
{agent_scratchpad}`,
      inputVariables: ["input", "agent_scratchpad"],
    }),
  }),
  allowedTools: ["search", "calculator"],
});

const result = await agent.invoke({
  input: `Who is Olivia Wilde's boyfriend? What is his current age raised to the 0.23 power?`,
});

Hierarchy

Constructors

constructor

Properties

ToolType llmChain outputParser

Accessors

allowedTools inputKeys returnValues

Methods

constructScratchPad finishToolName llmPrefix observationPrefix plan prepareForOutput returnStoppedResponse createPrompt deserialize fromLLMAndTools getDefaultOutputParser validateTools

Constructors

constructor

Properties

ToolType

ToolType: Tool

llmChain

llmChain: LLMChain<string, LLMType>

outputParser

outputParser: undefined | AgentActionOutputParser

Accessors

allowedTools

  • get allowedTools(): undefined | string[]

  • Returns undefined | string[]

inputKeys

  • get inputKeys(): string[]

  • Returns string[]

returnValues

  • get returnValues(): string[]

  • Returns string[]

Methods

constructScratchPad

  • Construct a scratchpad to let the agent continue its thought process

    Parameters

    • steps: AgentStep[]

    Returns Promise<string | BaseMessage[]>

finishToolName

  • Name of tool to use to terminate the chain.

    Returns string

llmPrefix

  • Prefix to append the LLM call with.

    Returns string

observationPrefix

  • Prefix to append the observation with.

    Returns string

plan

  • Decide what to do given some input.

    Parameters

    • steps: AgentStep[]

      Steps the LLM has taken so far, along with observations from each.

    • inputs: ChainValues

      User inputs.

    • Optional callbackManager: CallbackManager

      Callback manager to use for this call.

    Returns Promise<AgentAction | AgentFinish>

    Action specifying what tool to use.

prepareForOutput

  • Prepare the agent for output, if needed

    Parameters

    • _returnValues: Record<string, any>
    • _steps: AgentStep[]

    Returns Promise<Record<string, any>>

returnStoppedResponse

  • Return response when agent has been stopped due to max iterations

    Parameters

    • earlyStoppingMethod: StoppingMethod
    • steps: AgentStep[]
    • inputs: ChainValues
    • Optional callbackManager: CallbackManager

    Returns Promise<AgentFinish>

Static createPrompt

  • Create prompt in the style of the zero shot agent.

    Parameters

    • tools: Tool[]

      List of tools the agent will have access to, used to format the prompt.

    • Optional args: ZeroShotCreatePromptArgs

      Arguments to create the prompt with.

    Returns PromptTemplate<any, any>

Static deserialize

  • Load an agent from a json-like object describing it.

    Parameters

    • data: Object

    Returns Promise<ZeroShotAgent>

Static fromLLMAndTools

  • Creates a ZeroShotAgent from a Large Language Model and a set of tools.

    Parameters

    • llm: BaseLanguageModel<any, BaseLanguageModelCallOptions>

      The Large Language Model to use.

    • tools: Tool[]

      The tools for the agent to use.

    • Optional args: ZeroShotCreatePromptArgs & AgentArgs

      Optional arguments for creating the agent.

    Returns ZeroShotAgent

    A new instance of ZeroShotAgent.

Static getDefaultOutputParser

Static validateTools

  • Validates the tools for the ZeroShotAgent. Throws an error if any tool does not have a description.

    Parameters

    • tools: Tool[]

      List of tools to validate.

    Returns void

Settings

Member Visibility

Theme

On This Page

Generated using TypeDoc