@langchain/classic 1.0.17 → 1.0.18

package/dist/agents/react/prompt.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"prompt.cjs","names":[],"sources":["../../../src/agents/react/prompt.ts"],"sourcesContent":["export const FORMAT_INSTRUCTIONS = `Use the following format:\n\nQuestion: the input question you must answer\nThought: you should always think about what to do\nAction: the action to take, should be one of [{tool_names}]\nAction Input: the input to the action\nObservation: the result of the action\n... (this Thought/Action/Action Input/Observation can repeat N times)\nThought: I now know the final answer\nFinal Answer: the final answer to the original input question`;\n"],"mappings":";;AAAA,MAAa,sBAAsB,CAAC;;;;;;;;;6DASyB,CAAC"}
+{"version":3,"file":"prompt.cjs","names":[],"sources":["../../../src/agents/react/prompt.ts"],"sourcesContent":["export const FORMAT_INSTRUCTIONS = `Use the following format:\n\nQuestion: the input question you must answer\nThought: you should always think about what to do\nAction: the action to take, should be one of [{tool_names}]\nAction Input: the input to the action\nObservation: the result of the action\n... (this Thought/Action/Action Input/Observation can repeat N times)\nThought: I now know the final answer\nFinal Answer: the final answer to the original input question`;\n"],"mappings":";;AAAA,MAAa,sBAAsB"}

package/dist/agents/react/prompt.js.map

@@ -1 +1 @@
-{"version":3,"file":"prompt.js","names":[],"sources":["../../../src/agents/react/prompt.ts"],"sourcesContent":["export const FORMAT_INSTRUCTIONS = `Use the following format:\n\nQuestion: the input question you must answer\nThought: you should always think about what to do\nAction: the action to take, should be one of [{tool_names}]\nAction Input: the input to the action\nObservation: the result of the action\n... (this Thought/Action/Action Input/Observation can repeat N times)\nThought: I now know the final answer\nFinal Answer: the final answer to the original input question`;\n"],"mappings":";AAAA,MAAa,sBAAsB,CAAC;;;;;;;;;6DASyB,CAAC"}
+{"version":3,"file":"prompt.js","names":[],"sources":["../../../src/agents/react/prompt.ts"],"sourcesContent":["export const FORMAT_INSTRUCTIONS = `Use the following format:\n\nQuestion: the input question you must answer\nThought: you should always think about what to do\nAction: the action to take, should be one of [{tool_names}]\nAction Input: the input to the action\nObservation: the result of the action\n... (this Thought/Action/Action Input/Observation can repeat N times)\nThought: I now know the final answer\nFinal Answer: the final answer to the original input question`;\n"],"mappings":";AAAA,MAAa,sBAAsB"}

package/dist/agents/structured_chat/index.cjs

@@ -1,16 +1,16 @@
-const require_rolldown_runtime = require('../../_virtual/rolldown_runtime.cjs');
+const require_runtime = require('../../_virtual/_rolldown/runtime.cjs');
+const require_agents_format_scratchpad_log = require('../format_scratchpad/log.cjs');
 const require_agent = require('../agent.cjs');
 const require_llm_chain = require('../../chains/llm_chain.cjs');
 const require_prompt = require('./prompt.cjs');
 const require_outputParser = require('./outputParser.cjs');
 const require_tools_render = require('../../tools/render.cjs');
-const require_agents_format_scratchpad_log = require('../format_scratchpad/log.cjs');
-const __langchain_core_prompts = require_rolldown_runtime.__toESM(require("@langchain/core/prompts"));
-const __langchain_core_runnables = require_rolldown_runtime.__toESM(require("@langchain/core/runnables"));
-const __langchain_core_language_models_base = require_rolldown_runtime.__toESM(require("@langchain/core/language_models/base"));
-const __langchain_core_utils_function_calling = require_rolldown_runtime.__toESM(require("@langchain/core/utils/function_calling"));
-const __langchain_core_utils_types = require_rolldown_runtime.__toESM(require("@langchain/core/utils/types"));
-const __langchain_core_utils_json_schema = require_rolldown_runtime.__toESM(require("@langchain/core/utils/json_schema"));
+let _langchain_core_prompts = require("@langchain/core/prompts");
+let _langchain_core_runnables = require("@langchain/core/runnables");
+let _langchain_core_language_models_base = require("@langchain/core/language_models/base");
+let _langchain_core_utils_function_calling = require("@langchain/core/utils/function_calling");
+let _langchain_core_utils_types = require("@langchain/core/utils/types");
+let _langchain_core_utils_json_schema = require("@langchain/core/utils/json_schema");
 
 //#region src/agents/structured_chat/index.ts
 /**
@@ -86,7 +86,7 @@ var StructuredChatAgent = class StructuredChatAgent extends require_agent.Agent
 	*/
 	static createToolSchemasString(tools) {
 		return tools.map((tool) => {
-			const jsonSchema = (0, __langchain_core_utils_types.isInteropZodSchema)(tool.schema) ? (0, __langchain_core_utils_json_schema.toJsonSchema)(tool.schema) : tool.schema;
+			const jsonSchema = (0, _langchain_core_utils_types.isInteropZodSchema)(tool.schema) ? (0, _langchain_core_utils_json_schema.toJsonSchema)(tool.schema) : tool.schema;
 			return `${tool.name}: ${tool.description}, args: ${JSON.stringify(jsonSchema?.properties)}`;
 		}).join("\n");
 	}
@@ -102,14 +102,13 @@ var StructuredChatAgent = class StructuredChatAgent extends require_agent.Agent
 	*/
 	static createPrompt(tools, args) {
 		const { prefix = require_prompt.PREFIX, suffix = require_prompt.SUFFIX, inputVariables = ["input", "agent_scratchpad"], humanMessageTemplate = "{input}\n\n{agent_scratchpad}", memoryPrompts = [] } = args ?? {};
-		const template = [
-			prefix,
-			require_prompt.FORMAT_INSTRUCTIONS,
-			suffix
-		].join("\n\n");
 		const messages = [
-			new __langchain_core_prompts.SystemMessagePromptTemplate(new __langchain_core_prompts.PromptTemplate({
-				template,
+			new _langchain_core_prompts.SystemMessagePromptTemplate(new _langchain_core_prompts.PromptTemplate({
+				template: [
+					prefix,
+					require_prompt.FORMAT_INSTRUCTIONS,
+					suffix
+				].join("\n\n"),
 				inputVariables,
 				partialVariables: {
 					tool_schemas: StructuredChatAgent.createToolSchemasString(tools),
@@ -117,12 +116,12 @@ var StructuredChatAgent = class StructuredChatAgent extends require_agent.Agent
 				}
 			})),
 			...memoryPrompts,
-			new __langchain_core_prompts.HumanMessagePromptTemplate(new __langchain_core_prompts.PromptTemplate({
+			new _langchain_core_prompts.HumanMessagePromptTemplate(new _langchain_core_prompts.PromptTemplate({
 				template: humanMessageTemplate,
 				inputVariables
 			}))
 		];
-		return __langchain_core_prompts.ChatPromptTemplate.fromMessages(messages);
+		return _langchain_core_prompts.ChatPromptTemplate.fromMessages(messages);
 	}
 	/**
 	* Creates a StructuredChatAgent from an LLM and a list of tools.
@@ -140,13 +139,12 @@ var StructuredChatAgent = class StructuredChatAgent extends require_agent.Agent
 			llm,
 			toolNames: tools.map((tool) => tool.name)
 		});
-		const chain = new require_llm_chain.LLMChain({
-			prompt,
-			llm,
-			callbacks: args?.callbacks
-		});
 		return new StructuredChatAgent({
-			llmChain: chain,
+			llmChain: new require_llm_chain.LLMChain({
+				prompt,
+				llm,
+				callbacks: args?.callbacks
+			}),
 			outputParser,
 			allowedTools: tools.map((t) => t.name)
 		});
@@ -216,16 +214,16 @@ async function createStructuredChatAgent({ llm, tools, prompt, streamRunnable })
 	].filter((v) => !prompt.inputVariables.includes(v));
 	if (missingVariables.length > 0) throw new Error(`Provided prompt is missing required input variables: ${JSON.stringify(missingVariables)}`);
 	let toolNames = [];
-	if (tools.every(__langchain_core_language_models_base.isOpenAITool)) toolNames = tools.map((tool) => tool.function.name);
-	else if (tools.every(__langchain_core_utils_function_calling.isStructuredTool)) toolNames = tools.map((tool) => tool.name);
+	if (tools.every(_langchain_core_language_models_base.isOpenAITool)) toolNames = tools.map((tool) => tool.function.name);
+	else if (tools.every(_langchain_core_utils_function_calling.isStructuredTool)) toolNames = tools.map((tool) => tool.name);
 	else throw new Error("All tools must be either OpenAI or Structured tools, not a mix.");
 	const partialedPrompt = await prompt.partial({
 		tools: require_tools_render.renderTextDescriptionAndArgs(tools),
 		tool_names: toolNames.join(", ")
 	});
 	const llmWithStop = llm.withConfig({ stop: ["Observation"] });
-	const agent = require_agent.AgentRunnableSequence.fromRunnables([
-		__langchain_core_runnables.RunnablePassthrough.assign({ agent_scratchpad: (input) => require_agents_format_scratchpad_log.formatLogToString(input.steps) }),
+	return require_agent.AgentRunnableSequence.fromRunnables([
+		_langchain_core_runnables.RunnablePassthrough.assign({ agent_scratchpad: (input) => require_agents_format_scratchpad_log.formatLogToString(input.steps) }),
 		partialedPrompt,
 		llmWithStop,
 		require_outputParser.StructuredChatOutputParserWithRetries.fromLLM(llm, { toolNames })
@@ -234,7 +232,6 @@ async function createStructuredChatAgent({ llm, tools, prompt, streamRunnable })
 		streamRunnable,
 		singleAction: true
 	});
-	return agent;
 }
 
 //#endregion

package/dist/agents/structured_chat/index.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"index.cjs","names":["Agent","input: StructuredChatAgentInput","tools: StructuredToolInterface[]","fields?: OutputParserArgs & {\n      toolNames: string[];\n    }","StructuredChatOutputParserWithRetries","steps: AgentStep[]","args?: StructuredChatCreatePromptArgs","PREFIX","SUFFIX","FORMAT_INSTRUCTIONS","SystemMessagePromptTemplate","PromptTemplate","HumanMessagePromptTemplate","ChatPromptTemplate","llm: BaseLanguageModelInterface","args?: StructuredChatCreatePromptArgs & AgentArgs","LLMChain","toolNames: string[]","isOpenAITool","isStructuredTool","renderTextDescriptionAndArgs","AgentRunnableSequence","RunnablePassthrough","input: { steps: AgentStep[] }","formatLogToString"],"sources":["../../../src/agents/structured_chat/index.ts"],"sourcesContent":["import type { StructuredToolInterface } from \"@langchain/core/tools\";\nimport {\n  isOpenAITool,\n  type BaseLanguageModel,\n  type BaseLanguageModelInterface,\n  type ToolDefinition,\n} from \"@langchain/core/language_models/base\";\nimport { RunnablePassthrough } from \"@langchain/core/runnables\";\nimport type { BasePromptTemplate } from \"@langchain/core/prompts\";\nimport {\n  BaseMessagePromptTemplate,\n  ChatPromptTemplate,\n  HumanMessagePromptTemplate,\n  SystemMessagePromptTemplate,\n  PromptTemplate,\n} from \"@langchain/core/prompts\";\nimport { AgentStep } from \"@langchain/core/agents\";\nimport { isStructuredTool } from \"@langchain/core/utils/function_calling\";\nimport { isInteropZodSchema } from \"@langchain/core/utils/types\";\nimport {\n  type JsonSchema7Type,\n  toJsonSchema,\n} from \"@langchain/core/utils/json_schema\";\nimport { LLMChain } from \"../../chains/llm_chain.js\";\nimport { Optional } from \"../../types/type-utils.js\";\nimport {\n  Agent,\n  AgentArgs,\n  AgentRunnableSequence,\n  OutputParserArgs,\n} from \"../agent.js\";\nimport { AgentInput } from \"../types.js\";\nimport { StructuredChatOutputParserWithRetries } from \"./outputParser.js\";\nimport { FORMAT_INSTRUCTIONS, PREFIX, SUFFIX } from \"./prompt.js\";\nimport { renderTextDescriptionAndArgs } from \"../../tools/render.js\";\nimport { formatLogToString } from \"../format_scratchpad/log.js\";\n\n/**\n * Interface for arguments used to create a prompt for a\n * StructuredChatAgent.\n */\nexport interface StructuredChatCreatePromptArgs {\n  /** String to put after the list of tools. */\n  suffix?: string;\n  /** String to put before the list of tools. */\n  prefix?: string;\n  /** String to use directly as the human message template. */\n  humanMessageTemplate?: string;\n  /** List of input variables the final prompt will expect. */\n  inputVariables?: string[];\n  /** List of historical prompts from memory.  */\n  memoryPrompts?: BaseMessagePromptTemplate[];\n}\n\n/**\n * Type for input data for creating a StructuredChatAgent, with the\n * 'outputParser' property made optional.\n */\nexport type StructuredChatAgentInput = Optional<AgentInput, \"outputParser\">;\n\n/**\n * Agent that interoperates with Structured Tools using React logic.\n * @augments Agent\n */\nexport class StructuredChatAgent extends Agent {\n  static lc_name() {\n    return \"StructuredChatAgent\";\n  }\n\n  lc_namespace = [\"langchain\", \"agents\", \"structured_chat\"];\n\n  constructor(input: StructuredChatAgentInput) {\n    const outputParser =\n      input?.outputParser ?? StructuredChatAgent.getDefaultOutputParser();\n    super({ ...input, outputParser });\n  }\n\n  _agentType() {\n    return \"structured-chat-zero-shot-react-description\" as const;\n  }\n\n  observationPrefix() {\n    return \"Observation: \";\n  }\n\n  llmPrefix() {\n    return \"Thought:\";\n  }\n\n  _stop(): string[] {\n    return [\"Observation:\"];\n  }\n\n  /**\n   * Validates that all provided tools have a description. Throws an error\n   * if any tool lacks a description.\n   * @param tools Array of StructuredTool instances to validate.\n   */\n  static validateTools(tools: StructuredToolInterface[]) {\n    const descriptionlessTool = tools.find((tool) => !tool.description);\n    if (descriptionlessTool) {\n      const msg =\n        `Got a tool ${descriptionlessTool.name} without a description.` +\n        ` This agent requires descriptions for all tools.`;\n      throw new Error(msg);\n    }\n  }\n\n  /**\n   * Returns a default output parser for the StructuredChatAgent. If an LLM\n   * is provided, it creates an output parser with retry logic from the LLM.\n   * @param fields Optional fields to customize the output parser. Can include an LLM and a list of tool names.\n   * @returns An instance of StructuredChatOutputParserWithRetries.\n   */\n  static getDefaultOutputParser(\n    fields?: OutputParserArgs & {\n      toolNames: string[];\n    }\n  ) {\n    if (fields?.llm) {\n      return StructuredChatOutputParserWithRetries.fromLLM(fields.llm, {\n        toolNames: fields.toolNames,\n      });\n    }\n    return new StructuredChatOutputParserWithRetries({\n      toolNames: fields?.toolNames,\n    });\n  }\n\n  /**\n   * Constructs the agent's scratchpad from a list of steps. If the agent's\n   * scratchpad is not empty, it prepends a message indicating that the\n   * agent has not seen any previous work.\n   * @param steps Array of AgentStep instances to construct the scratchpad from.\n   * @returns A Promise that resolves to a string representing the agent's scratchpad.\n   */\n  async constructScratchPad(steps: AgentStep[]): Promise<string> {\n    const agentScratchpad = await super.constructScratchPad(steps);\n    if (agentScratchpad) {\n      return `This was your previous work (but I haven't seen any of it! I only see what you return as final answer):\\n${agentScratchpad}`;\n    }\n    return agentScratchpad;\n  }\n\n  /**\n   * Creates a string representation of the schemas of the provided tools.\n   * @param tools Array of StructuredTool instances to create the schemas string from.\n   * @returns A string representing the schemas of the provided tools.\n   */\n  static createToolSchemasString(tools: StructuredToolInterface[]) {\n    return tools\n      .map((tool) => {\n        const jsonSchema = (\n          isInteropZodSchema(tool.schema)\n            ? toJsonSchema(tool.schema)\n            : tool.schema\n        ) as { properties?: Record<string, JsonSchema7Type> } | undefined;\n        return `${tool.name}: ${tool.description}, args: ${JSON.stringify(\n          jsonSchema?.properties\n        )}`;\n      })\n      .join(\"\\n\");\n  }\n\n  /**\n   * Create prompt in the style of the agent.\n   *\n   * @param tools - List of tools the agent will have access to, used to format the prompt.\n   * @param args - Arguments to create the prompt with.\n   * @param args.suffix - String to put after the list of tools.\n   * @param args.prefix - String to put before the list of tools.\n   * @param args.inputVariables List of input variables the final prompt will expect.\n   * @param args.memoryPrompts List of historical prompts from memory.\n   */\n  static createPrompt(\n    tools: StructuredToolInterface[],\n    args?: StructuredChatCreatePromptArgs\n  ) {\n    const {\n      prefix = PREFIX,\n      suffix = SUFFIX,\n      inputVariables = [\"input\", \"agent_scratchpad\"],\n      humanMessageTemplate = \"{input}\\n\\n{agent_scratchpad}\",\n      memoryPrompts = [],\n    } = args ?? {};\n    const template = [prefix, FORMAT_INSTRUCTIONS, suffix].join(\"\\n\\n\");\n    const messages = [\n      new SystemMessagePromptTemplate(\n        new PromptTemplate({\n          template,\n          inputVariables,\n          partialVariables: {\n            tool_schemas: StructuredChatAgent.createToolSchemasString(tools),\n            tool_names: tools.map((tool) => tool.name).join(\", \"),\n          },\n        })\n      ),\n      ...memoryPrompts,\n      new HumanMessagePromptTemplate(\n        new PromptTemplate({\n          template: humanMessageTemplate,\n          inputVariables,\n        })\n      ),\n    ];\n    return ChatPromptTemplate.fromMessages(messages);\n  }\n\n  /**\n   * Creates a StructuredChatAgent from an LLM and a list of tools.\n   * Validates the tools, creates a prompt, and sets up an LLM chain for the\n   * agent.\n   * @param llm BaseLanguageModel instance to create the agent from.\n   * @param tools Array of StructuredTool instances to create the agent from.\n   * @param args Optional arguments to customize the creation of the agent. Can include arguments for creating the prompt and AgentArgs.\n   * @returns A new instance of StructuredChatAgent.\n   */\n  static fromLLMAndTools(\n    llm: BaseLanguageModelInterface,\n    tools: StructuredToolInterface[],\n    args?: StructuredChatCreatePromptArgs & AgentArgs\n  ) {\n    StructuredChatAgent.validateTools(tools);\n    const prompt = StructuredChatAgent.createPrompt(tools, args);\n    const outputParser =\n      args?.outputParser ??\n      StructuredChatAgent.getDefaultOutputParser({\n        llm,\n        toolNames: tools.map((tool) => tool.name),\n      });\n    const chain = new LLMChain({\n      prompt,\n      llm,\n      callbacks: args?.callbacks,\n    });\n\n    return new StructuredChatAgent({\n      llmChain: chain,\n      outputParser,\n      allowedTools: tools.map((t) => t.name),\n    });\n  }\n}\n\n/**\n * Params used by the createStructuredChatAgent function.\n */\nexport type CreateStructuredChatAgentParams = {\n  /** LLM to use as the agent. */\n  llm: BaseLanguageModelInterface;\n  /** Tools this agent has access to. */\n  tools: (StructuredToolInterface | ToolDefinition)[];\n  /**\n   * The prompt to use. Must have input keys for\n   * `tools`, `tool_names`, and `agent_scratchpad`.\n   */\n  prompt: BasePromptTemplate;\n  /**\n   * Whether to invoke the underlying model in streaming mode,\n   * allowing streaming of intermediate steps. Defaults to true.\n   */\n  streamRunnable?: boolean;\n};\n\n/**\n * Create an agent aimed at supporting tools with multiple inputs.\n * @param params Params required to create the agent. Includes an LLM, tools, and prompt.\n * @returns A runnable sequence representing an agent. It takes as input all the same input\n *     variables as the prompt passed in does. It returns as output either an\n *     AgentAction or AgentFinish.\n *\n * @example\n * ```typescript\n * import { AgentExecutor, createStructuredChatAgent } from \"langchain/agents\";\n * import { pull } from \"langchain/hub\";\n * import type { ChatPromptTemplate } from \"@langchain/core/prompts\";\n * import { AIMessage, HumanMessage } from \"@langchain/core/messages\";\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n *\n * // Define the tools the agent will have access to.\n * const tools = [...];\n *\n * // Get the prompt to use - you can modify this!\n * // If you want to see the prompt in full, you can at:\n * // https://smith.langchain.com/hub/hwchase17/structured-chat-agent\n * const prompt = await pull<ChatPromptTemplate>(\n *   \"hwchase17/structured-chat-agent\"\n * );\n *\n * const llm = new ChatOpenAI({\n *   temperature: 0,\n *   model: \"gpt-3.5-turbo-1106\",\n * });\n *\n * const agent = await createStructuredChatAgent({\n *   llm,\n *   tools,\n *   prompt,\n * });\n *\n * const agentExecutor = new AgentExecutor({\n *   agent,\n *   tools,\n * });\n *\n * const result = await agentExecutor.invoke({\n *   input: \"what is LangChain?\",\n * });\n *\n * // With chat history\n * const result2 = await agentExecutor.invoke({\n *   input: \"what's my name?\",\n *   chat_history: [\n *     new HumanMessage(\"hi! my name is cob\"),\n *     new AIMessage(\"Hello Cob! How can I assist you today?\"),\n *   ],\n * });\n * ```\n */\nexport async function createStructuredChatAgent({\n  llm,\n  tools,\n  prompt,\n  streamRunnable,\n}: CreateStructuredChatAgentParams) {\n  const missingVariables = [\"tools\", \"tool_names\", \"agent_scratchpad\"].filter(\n    (v) => !prompt.inputVariables.includes(v)\n  );\n  if (missingVariables.length > 0) {\n    throw new Error(\n      `Provided prompt is missing required input variables: ${JSON.stringify(\n        missingVariables\n      )}`\n    );\n  }\n  let toolNames: string[] = [];\n  if (tools.every(isOpenAITool)) {\n    toolNames = tools.map((tool) => tool.function.name);\n  } else if (tools.every(isStructuredTool)) {\n    toolNames = tools.map((tool) => tool.name);\n  } else {\n    throw new Error(\n      \"All tools must be either OpenAI or Structured tools, not a mix.\"\n    );\n  }\n  const partialedPrompt = await prompt.partial({\n    tools: renderTextDescriptionAndArgs(tools),\n    tool_names: toolNames.join(\", \"),\n  });\n  // TODO: Add .bind to core runnable interface.\n  const llmWithStop = (llm as BaseLanguageModel).withConfig({\n    stop: [\"Observation\"],\n  });\n  const agent = AgentRunnableSequence.fromRunnables(\n    [\n      RunnablePassthrough.assign({\n        agent_scratchpad: (input: { steps: AgentStep[] }) =>\n          formatLogToString(input.steps),\n      }),\n      partialedPrompt,\n      llmWithStop,\n      StructuredChatOutputParserWithRetries.fromLLM(llm, {\n        toolNames,\n      }),\n    ],\n    {\n      name: \"StructuredChatAgent\",\n      streamRunnable,\n      singleAction: true,\n    }\n  );\n  return agent;\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;AAgEA,IAAa,sBAAb,MAAa,4BAA4BA,oBAAM;CAC7C,OAAO,UAAU;AACf,SAAO;CACR;CAED,eAAe;EAAC;EAAa;EAAU;CAAkB;CAEzD,YAAYC,OAAiC;EAC3C,MAAM,eACJ,OAAO,gBAAgB,oBAAoB,wBAAwB;EACrE,MAAM;GAAE,GAAG;GAAO;EAAc,EAAC;CAClC;CAED,aAAa;AACX,SAAO;CACR;CAED,oBAAoB;AAClB,SAAO;CACR;CAED,YAAY;AACV,SAAO;CACR;CAED,QAAkB;AAChB,SAAO,CAAC,cAAe;CACxB;;;;;;CAOD,OAAO,cAAcC,OAAkC;EACrD,MAAM,sBAAsB,MAAM,KAAK,CAAC,SAAS,CAAC,KAAK,YAAY;AACnE,MAAI,qBAAqB;GACvB,MAAM,MACJ,CAAC,WAAW,EAAE,oBAAoB,KAAK,uEAAuB,CACZ;AACpD,SAAM,IAAI,MAAM;EACjB;CACF;;;;;;;CAQD,OAAO,uBACLC,QAGA;AACA,MAAI,QAAQ,IACV,QAAOC,2DAAsC,QAAQ,OAAO,KAAK,EAC/D,WAAW,OAAO,UACnB,EAAC;AAEJ,SAAO,IAAIA,2DAAsC,EAC/C,WAAW,QAAQ,UACpB;CACF;;;;;;;;CASD,MAAM,oBAAoBC,OAAqC;EAC7D,MAAM,kBAAkB,MAAM,MAAM,oBAAoB,MAAM;AAC9D,MAAI,gBACF,QAAO,CAAC,yGAAyG,EAAE,iBAAiB;AAEtI,SAAO;CACR;;;;;;CAOD,OAAO,wBAAwBH,OAAkC;AAC/D,SAAO,MACJ,IAAI,CAAC,SAAS;GACb,MAAM,kEACe,KAAK,OAAO,wDACd,KAAK,OAAO,GACzB,KAAK;AAEX,UAAO,GAAG,KAAK,KAAK,EAAE,EAAE,KAAK,YAAY,QAAQ,EAAE,KAAK,UACtD,YAAY,WACb,EAAE;EACJ,EAAC,CACD,KAAK,KAAK;CACd;;;;;;;;;;;CAYD,OAAO,aACLA,OACAI,MACA;EACA,MAAM,EACJ,SAASC,uBACT,SAASC,uBACT,iBAAiB,CAAC,SAAS,kBAAmB,GAC9C,uBAAuB,iCACvB,gBAAgB,CAAE,GACnB,GAAG,QAAQ,CAAE;EACd,MAAM,WAAW;GAAC;GAAQC;GAAqB;EAAO,EAAC,KAAK,OAAO;EACnE,MAAM,WAAW;GACf,IAAIC,qDACF,IAAIC,wCAAe;IACjB;IACA;IACA,kBAAkB;KAChB,cAAc,oBAAoB,wBAAwB,MAAM;KAChE,YAAY,MAAM,IAAI,CAAC,SAAS,KAAK,KAAK,CAAC,KAAK,KAAK;IACtD;GACF;GAEH,GAAG;GACH,IAAIC,oDACF,IAAID,wCAAe;IACjB,UAAU;IACV;GACD;EAEJ;AACD,SAAOE,4CAAmB,aAAa,SAAS;CACjD;;;;;;;;;;CAWD,OAAO,gBACLC,KACAZ,OACAa,MACA;EACA,oBAAoB,cAAc,MAAM;EACxC,MAAM,SAAS,oBAAoB,aAAa,OAAO,KAAK;EAC5D,MAAM,eACJ,MAAM,gBACN,oBAAoB,uBAAuB;GACzC;GACA,WAAW,MAAM,IAAI,CAAC,SAAS,KAAK,KAAK;EAC1C,EAAC;EACJ,MAAM,QAAQ,IAAIC,2BAAS;GACzB;GACA;GACA,WAAW,MAAM;EAClB;AAED,SAAO,IAAI,oBAAoB;GAC7B,UAAU;GACV;GACA,cAAc,MAAM,IAAI,CAAC,MAAM,EAAE,KAAK;EACvC;CACF;AACF;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AA8ED,eAAsB,0BAA0B,EAC9C,KACA,OACA,QACA,gBACgC,EAAE;CAClC,MAAM,mBAAmB;EAAC;EAAS;EAAc;CAAmB,EAAC,OACnE,CAAC,MAAM,CAAC,OAAO,eAAe,SAAS,EAAE,CAC1C;AACD,KAAI,iBAAiB,SAAS,EAC5B,OAAM,IAAI,MACR,CAAC,qDAAqD,EAAE,KAAK,UAC3D,iBACD,EAAE;CAGP,IAAIC,YAAsB,CAAE;AAC5B,KAAI,MAAM,MAAMC,mDAAa,EAC3B,YAAY,MAAM,IAAI,CAAC,SAAS,KAAK,SAAS,KAAK;UAC1C,MAAM,MAAMC,yDAAiB,EACtC,YAAY,MAAM,IAAI,CAAC,SAAS,KAAK,KAAK;KAE1C,OAAM,IAAI,MACR;CAGJ,MAAM,kBAAkB,MAAM,OAAO,QAAQ;EAC3C,OAAOC,kDAA6B,MAAM;EAC1C,YAAY,UAAU,KAAK,KAAK;CACjC,EAAC;CAEF,MAAM,cAAe,IAA0B,WAAW,EACxD,MAAM,CAAC,aAAc,EACtB,EAAC;CACF,MAAM,QAAQC,oCAAsB,cAClC;EACEC,+CAAoB,OAAO,EACzB,kBAAkB,CAACC,UACjBC,uDAAkB,MAAM,MAAM,CACjC,EAAC;EACF;EACA;EACApB,2DAAsC,QAAQ,KAAK,EACjD,UACD,EAAC;CACH,GACD;EACE,MAAM;EACN;EACA,cAAc;CACf,EACF;AACD,QAAO;AACR"}
+{"version":3,"file":"index.cjs","names":["Agent","StructuredChatOutputParserWithRetries","PREFIX","SUFFIX","SystemMessagePromptTemplate","PromptTemplate","FORMAT_INSTRUCTIONS","HumanMessagePromptTemplate","ChatPromptTemplate","LLMChain","isOpenAITool","isStructuredTool","renderTextDescriptionAndArgs","AgentRunnableSequence","RunnablePassthrough","formatLogToString"],"sources":["../../../src/agents/structured_chat/index.ts"],"sourcesContent":["import type { StructuredToolInterface } from \"@langchain/core/tools\";\nimport {\n  isOpenAITool,\n  type BaseLanguageModel,\n  type BaseLanguageModelInterface,\n  type ToolDefinition,\n} from \"@langchain/core/language_models/base\";\nimport { RunnablePassthrough } from \"@langchain/core/runnables\";\nimport type { BasePromptTemplate } from \"@langchain/core/prompts\";\nimport {\n  BaseMessagePromptTemplate,\n  ChatPromptTemplate,\n  HumanMessagePromptTemplate,\n  SystemMessagePromptTemplate,\n  PromptTemplate,\n} from \"@langchain/core/prompts\";\nimport { AgentStep } from \"@langchain/core/agents\";\nimport { isStructuredTool } from \"@langchain/core/utils/function_calling\";\nimport { isInteropZodSchema } from \"@langchain/core/utils/types\";\nimport {\n  type JsonSchema7Type,\n  toJsonSchema,\n} from \"@langchain/core/utils/json_schema\";\nimport { LLMChain } from \"../../chains/llm_chain.js\";\nimport { Optional } from \"../../types/type-utils.js\";\nimport {\n  Agent,\n  AgentArgs,\n  AgentRunnableSequence,\n  OutputParserArgs,\n} from \"../agent.js\";\nimport { AgentInput } from \"../types.js\";\nimport { StructuredChatOutputParserWithRetries } from \"./outputParser.js\";\nimport { FORMAT_INSTRUCTIONS, PREFIX, SUFFIX } from \"./prompt.js\";\nimport { renderTextDescriptionAndArgs } from \"../../tools/render.js\";\nimport { formatLogToString } from \"../format_scratchpad/log.js\";\n\n/**\n * Interface for arguments used to create a prompt for a\n * StructuredChatAgent.\n */\nexport interface StructuredChatCreatePromptArgs {\n  /** String to put after the list of tools. */\n  suffix?: string;\n  /** String to put before the list of tools. */\n  prefix?: string;\n  /** String to use directly as the human message template. */\n  humanMessageTemplate?: string;\n  /** List of input variables the final prompt will expect. */\n  inputVariables?: string[];\n  /** List of historical prompts from memory.  */\n  memoryPrompts?: BaseMessagePromptTemplate[];\n}\n\n/**\n * Type for input data for creating a StructuredChatAgent, with the\n * 'outputParser' property made optional.\n */\nexport type StructuredChatAgentInput = Optional<AgentInput, \"outputParser\">;\n\n/**\n * Agent that interoperates with Structured Tools using React logic.\n * @augments Agent\n */\nexport class StructuredChatAgent extends Agent {\n  static lc_name() {\n    return \"StructuredChatAgent\";\n  }\n\n  lc_namespace = [\"langchain\", \"agents\", \"structured_chat\"];\n\n  constructor(input: StructuredChatAgentInput) {\n    const outputParser =\n      input?.outputParser ?? StructuredChatAgent.getDefaultOutputParser();\n    super({ ...input, outputParser });\n  }\n\n  _agentType() {\n    return \"structured-chat-zero-shot-react-description\" as const;\n  }\n\n  observationPrefix() {\n    return \"Observation: \";\n  }\n\n  llmPrefix() {\n    return \"Thought:\";\n  }\n\n  _stop(): string[] {\n    return [\"Observation:\"];\n  }\n\n  /**\n   * Validates that all provided tools have a description. Throws an error\n   * if any tool lacks a description.\n   * @param tools Array of StructuredTool instances to validate.\n   */\n  static validateTools(tools: StructuredToolInterface[]) {\n    const descriptionlessTool = tools.find((tool) => !tool.description);\n    if (descriptionlessTool) {\n      const msg =\n        `Got a tool ${descriptionlessTool.name} without a description.` +\n        ` This agent requires descriptions for all tools.`;\n      throw new Error(msg);\n    }\n  }\n\n  /**\n   * Returns a default output parser for the StructuredChatAgent. If an LLM\n   * is provided, it creates an output parser with retry logic from the LLM.\n   * @param fields Optional fields to customize the output parser. Can include an LLM and a list of tool names.\n   * @returns An instance of StructuredChatOutputParserWithRetries.\n   */\n  static getDefaultOutputParser(\n    fields?: OutputParserArgs & {\n      toolNames: string[];\n    }\n  ) {\n    if (fields?.llm) {\n      return StructuredChatOutputParserWithRetries.fromLLM(fields.llm, {\n        toolNames: fields.toolNames,\n      });\n    }\n    return new StructuredChatOutputParserWithRetries({\n      toolNames: fields?.toolNames,\n    });\n  }\n\n  /**\n   * Constructs the agent's scratchpad from a list of steps. If the agent's\n   * scratchpad is not empty, it prepends a message indicating that the\n   * agent has not seen any previous work.\n   * @param steps Array of AgentStep instances to construct the scratchpad from.\n   * @returns A Promise that resolves to a string representing the agent's scratchpad.\n   */\n  async constructScratchPad(steps: AgentStep[]): Promise<string> {\n    const agentScratchpad = await super.constructScratchPad(steps);\n    if (agentScratchpad) {\n      return `This was your previous work (but I haven't seen any of it! I only see what you return as final answer):\\n${agentScratchpad}`;\n    }\n    return agentScratchpad;\n  }\n\n  /**\n   * Creates a string representation of the schemas of the provided tools.\n   * @param tools Array of StructuredTool instances to create the schemas string from.\n   * @returns A string representing the schemas of the provided tools.\n   */\n  static createToolSchemasString(tools: StructuredToolInterface[]) {\n    return tools\n      .map((tool) => {\n        const jsonSchema = (\n          isInteropZodSchema(tool.schema)\n            ? toJsonSchema(tool.schema)\n            : tool.schema\n        ) as { properties?: Record<string, JsonSchema7Type> } | undefined;\n        return `${tool.name}: ${tool.description}, args: ${JSON.stringify(\n          jsonSchema?.properties\n        )}`;\n      })\n      .join(\"\\n\");\n  }\n\n  /**\n   * Create prompt in the style of the agent.\n   *\n   * @param tools - List of tools the agent will have access to, used to format the prompt.\n   * @param args - Arguments to create the prompt with.\n   * @param args.suffix - String to put after the list of tools.\n   * @param args.prefix - String to put before the list of tools.\n   * @param args.inputVariables List of input variables the final prompt will expect.\n   * @param args.memoryPrompts List of historical prompts from memory.\n   */\n  static createPrompt(\n    tools: StructuredToolInterface[],\n    args?: StructuredChatCreatePromptArgs\n  ) {\n    const {\n      prefix = PREFIX,\n      suffix = SUFFIX,\n      inputVariables = [\"input\", \"agent_scratchpad\"],\n      humanMessageTemplate = \"{input}\\n\\n{agent_scratchpad}\",\n      memoryPrompts = [],\n    } = args ?? {};\n    const template = [prefix, FORMAT_INSTRUCTIONS, suffix].join(\"\\n\\n\");\n    const messages = [\n      new SystemMessagePromptTemplate(\n        new PromptTemplate({\n          template,\n          inputVariables,\n          partialVariables: {\n            tool_schemas: StructuredChatAgent.createToolSchemasString(tools),\n            tool_names: tools.map((tool) => tool.name).join(\", \"),\n          },\n        })\n      ),\n      ...memoryPrompts,\n      new HumanMessagePromptTemplate(\n        new PromptTemplate({\n          template: humanMessageTemplate,\n          inputVariables,\n        })\n      ),\n    ];\n    return ChatPromptTemplate.fromMessages(messages);\n  }\n\n  /**\n   * Creates a StructuredChatAgent from an LLM and a list of tools.\n   * Validates the tools, creates a prompt, and sets up an LLM chain for the\n   * agent.\n   * @param llm BaseLanguageModel instance to create the agent from.\n   * @param tools Array of StructuredTool instances to create the agent from.\n   * @param args Optional arguments to customize the creation of the agent. Can include arguments for creating the prompt and AgentArgs.\n   * @returns A new instance of StructuredChatAgent.\n   */\n  static fromLLMAndTools(\n    llm: BaseLanguageModelInterface,\n    tools: StructuredToolInterface[],\n    args?: StructuredChatCreatePromptArgs & AgentArgs\n  ) {\n    StructuredChatAgent.validateTools(tools);\n    const prompt = StructuredChatAgent.createPrompt(tools, args);\n    const outputParser =\n      args?.outputParser ??\n      StructuredChatAgent.getDefaultOutputParser({\n        llm,\n        toolNames: tools.map((tool) => tool.name),\n      });\n    const chain = new LLMChain({\n      prompt,\n      llm,\n      callbacks: args?.callbacks,\n    });\n\n    return new StructuredChatAgent({\n      llmChain: chain,\n      outputParser,\n      allowedTools: tools.map((t) => t.name),\n    });\n  }\n}\n\n/**\n * Params used by the createStructuredChatAgent function.\n */\nexport type CreateStructuredChatAgentParams = {\n  /** LLM to use as the agent. */\n  llm: BaseLanguageModelInterface;\n  /** Tools this agent has access to. */\n  tools: (StructuredToolInterface | ToolDefinition)[];\n  /**\n   * The prompt to use. Must have input keys for\n   * `tools`, `tool_names`, and `agent_scratchpad`.\n   */\n  prompt: BasePromptTemplate;\n  /**\n   * Whether to invoke the underlying model in streaming mode,\n   * allowing streaming of intermediate steps. Defaults to true.\n   */\n  streamRunnable?: boolean;\n};\n\n/**\n * Create an agent aimed at supporting tools with multiple inputs.\n * @param params Params required to create the agent. Includes an LLM, tools, and prompt.\n * @returns A runnable sequence representing an agent. It takes as input all the same input\n *     variables as the prompt passed in does. It returns as output either an\n *     AgentAction or AgentFinish.\n *\n * @example\n * ```typescript\n * import { AgentExecutor, createStructuredChatAgent } from \"langchain/agents\";\n * import { pull } from \"langchain/hub\";\n * import type { ChatPromptTemplate } from \"@langchain/core/prompts\";\n * import { AIMessage, HumanMessage } from \"@langchain/core/messages\";\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n *\n * // Define the tools the agent will have access to.\n * const tools = [...];\n *\n * // Get the prompt to use - you can modify this!\n * // If you want to see the prompt in full, you can at:\n * // https://smith.langchain.com/hub/hwchase17/structured-chat-agent\n * const prompt = await pull<ChatPromptTemplate>(\n *   \"hwchase17/structured-chat-agent\"\n * );\n *\n * const llm = new ChatOpenAI({\n *   temperature: 0,\n *   model: \"gpt-3.5-turbo-1106\",\n * });\n *\n * const agent = await createStructuredChatAgent({\n *   llm,\n *   tools,\n *   prompt,\n * });\n *\n * const agentExecutor = new AgentExecutor({\n *   agent,\n *   tools,\n * });\n *\n * const result = await agentExecutor.invoke({\n *   input: \"what is LangChain?\",\n * });\n *\n * // With chat history\n * const result2 = await agentExecutor.invoke({\n *   input: \"what's my name?\",\n *   chat_history: [\n *     new HumanMessage(\"hi! my name is cob\"),\n *     new AIMessage(\"Hello Cob! How can I assist you today?\"),\n *   ],\n * });\n * ```\n */\nexport async function createStructuredChatAgent({\n  llm,\n  tools,\n  prompt,\n  streamRunnable,\n}: CreateStructuredChatAgentParams) {\n  const missingVariables = [\"tools\", \"tool_names\", \"agent_scratchpad\"].filter(\n    (v) => !prompt.inputVariables.includes(v)\n  );\n  if (missingVariables.length > 0) {\n    throw new Error(\n      `Provided prompt is missing required input variables: ${JSON.stringify(\n        missingVariables\n      )}`\n    );\n  }\n  let toolNames: string[] = [];\n  if (tools.every(isOpenAITool)) {\n    toolNames = tools.map((tool) => tool.function.name);\n  } else if (tools.every(isStructuredTool)) {\n    toolNames = tools.map((tool) => tool.name);\n  } else {\n    throw new Error(\n      \"All tools must be either OpenAI or Structured tools, not a mix.\"\n    );\n  }\n  const partialedPrompt = await prompt.partial({\n    tools: renderTextDescriptionAndArgs(tools),\n    tool_names: toolNames.join(\", \"),\n  });\n  // TODO: Add .bind to core runnable interface.\n  const llmWithStop = (llm as BaseLanguageModel).withConfig({\n    stop: [\"Observation\"],\n  });\n  const agent = AgentRunnableSequence.fromRunnables(\n    [\n      RunnablePassthrough.assign({\n        agent_scratchpad: (input: { steps: AgentStep[] }) =>\n          formatLogToString(input.steps),\n      }),\n      partialedPrompt,\n      llmWithStop,\n      StructuredChatOutputParserWithRetries.fromLLM(llm, {\n        toolNames,\n      }),\n    ],\n    {\n      name: \"StructuredChatAgent\",\n      streamRunnable,\n      singleAction: true,\n    }\n  );\n  return agent;\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;AAgEA,IAAa,sBAAb,MAAa,4BAA4BA,oBAAM;CAC7C,OAAO,UAAU;AACf,SAAO;;CAGT,eAAe;EAAC;EAAa;EAAU;EAAkB;CAEzD,YAAY,OAAiC;EAC3C,MAAM,eACJ,OAAO,gBAAgB,oBAAoB,wBAAwB;AACrE,QAAM;GAAE,GAAG;GAAO;GAAc,CAAC;;CAGnC,aAAa;AACX,SAAO;;CAGT,oBAAoB;AAClB,SAAO;;CAGT,YAAY;AACV,SAAO;;CAGT,QAAkB;AAChB,SAAO,CAAC,eAAe;;;;;;;CAQzB,OAAO,cAAc,OAAkC;EACrD,MAAM,sBAAsB,MAAM,MAAM,SAAS,CAAC,KAAK,YAAY;AACnE,MAAI,qBAAqB;GACvB,MAAM,MACJ,cAAc,oBAAoB,KAAK;AAEzC,SAAM,IAAI,MAAM,IAAI;;;;;;;;;CAUxB,OAAO,uBACL,QAGA;AACA,MAAI,QAAQ,IACV,QAAOC,2DAAsC,QAAQ,OAAO,KAAK,EAC/D,WAAW,OAAO,WACnB,CAAC;AAEJ,SAAO,IAAIA,2DAAsC,EAC/C,WAAW,QAAQ,WACpB,CAAC;;;;;;;;;CAUJ,MAAM,oBAAoB,OAAqC;EAC7D,MAAM,kBAAkB,MAAM,MAAM,oBAAoB,MAAM;AAC9D,MAAI,gBACF,QAAO,4GAA4G;AAErH,SAAO;;;;;;;CAQT,OAAO,wBAAwB,OAAkC;AAC/D,SAAO,MACJ,KAAK,SAAS;GACb,MAAM,iEACe,KAAK,OAAO,uDACd,KAAK,OAAO,GACzB,KAAK;AAEX,UAAO,GAAG,KAAK,KAAK,IAAI,KAAK,YAAY,UAAU,KAAK,UACtD,YAAY,WACb;IACD,CACD,KAAK,KAAK;;;;;;;;;;;;CAaf,OAAO,aACL,OACA,MACA;EACA,MAAM,EACJ,SAASC,uBACT,SAASC,uBACT,iBAAiB,CAAC,SAAS,mBAAmB,EAC9C,uBAAuB,iCACvB,gBAAgB,EAAE,KAChB,QAAQ,EAAE;EAEd,MAAM,WAAW;GACf,IAAIC,oDACF,IAAIC,uCAAe;IACjB,UAJW;KAAC;KAAQC;KAAqB;KAAO,CAAC,KAAK,OAAO;IAK7D;IACA,kBAAkB;KAChB,cAAc,oBAAoB,wBAAwB,MAAM;KAChE,YAAY,MAAM,KAAK,SAAS,KAAK,KAAK,CAAC,KAAK,KAAK;KACtD;IACF,CAAC,CACH;GACD,GAAG;GACH,IAAIC,mDACF,IAAIF,uCAAe;IACjB,UAAU;IACV;IACD,CAAC,CACH;GACF;AACD,SAAOG,2CAAmB,aAAa,SAAS;;;;;;;;;;;CAYlD,OAAO,gBACL,KACA,OACA,MACA;AACA,sBAAoB,cAAc,MAAM;EACxC,MAAM,SAAS,oBAAoB,aAAa,OAAO,KAAK;EAC5D,MAAM,eACJ,MAAM,gBACN,oBAAoB,uBAAuB;GACzC;GACA,WAAW,MAAM,KAAK,SAAS,KAAK,KAAK;GAC1C,CAAC;AAOJ,SAAO,IAAI,oBAAoB;GAC7B,UAPY,IAAIC,2BAAS;IACzB;IACA;IACA,WAAW,MAAM;IAClB,CAAC;GAIA;GACA,cAAc,MAAM,KAAK,MAAM,EAAE,KAAK;GACvC,CAAC;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAgFN,eAAsB,0BAA0B,EAC9C,KACA,OACA,QACA,kBACkC;CAClC,MAAM,mBAAmB;EAAC;EAAS;EAAc;EAAmB,CAAC,QAClE,MAAM,CAAC,OAAO,eAAe,SAAS,EAAE,CAC1C;AACD,KAAI,iBAAiB,SAAS,EAC5B,OAAM,IAAI,MACR,wDAAwD,KAAK,UAC3D,iBACD,GACF;CAEH,IAAI,YAAsB,EAAE;AAC5B,KAAI,MAAM,MAAMC,kDAAa,CAC3B,aAAY,MAAM,KAAK,SAAS,KAAK,SAAS,KAAK;UAC1C,MAAM,MAAMC,wDAAiB,CACtC,aAAY,MAAM,KAAK,SAAS,KAAK,KAAK;KAE1C,OAAM,IAAI,MACR,kEACD;CAEH,MAAM,kBAAkB,MAAM,OAAO,QAAQ;EAC3C,OAAOC,kDAA6B,MAAM;EAC1C,YAAY,UAAU,KAAK,KAAK;EACjC,CAAC;CAEF,MAAM,cAAe,IAA0B,WAAW,EACxD,MAAM,CAAC,cAAc,EACtB,CAAC;AAmBF,QAlBcC,oCAAsB,cAClC;EACEC,8CAAoB,OAAO,EACzB,mBAAmB,UACjBC,uDAAkB,MAAM,MAAM,EACjC,CAAC;EACF;EACA;EACAd,2DAAsC,QAAQ,KAAK,EACjD,WACD,CAAC;EACH,EACD;EACE,MAAM;EACN;EACA,cAAc;EACf,CACF"}

package/dist/agents/structured_chat/index.d.cts

@@ -9,7 +9,6 @@ import { BaseLanguageModelInterface, ToolDefinition } from "@langchain/core/lang
 import { StructuredToolInterface } from "@langchain/core/tools";
 
 //#region src/agents/structured_chat/index.d.ts
-
 /**
  * Interface for arguments used to create a prompt for a
  * StructuredChatAgent.
@@ -98,9 +97,7 @@ declare class StructuredChatAgent extends Agent {
  * Params used by the createStructuredChatAgent function.
  */
 type CreateStructuredChatAgentParams = {
-  /** LLM to use as the agent. */
-  llm: BaseLanguageModelInterface;
-  /** Tools this agent has access to. */
+  /** LLM to use as the agent. */llm: BaseLanguageModelInterface; /** Tools this agent has access to. */
   tools: (StructuredToolInterface | ToolDefinition)[];
   /**
    * The prompt to use. Must have input keys for

package/dist/agents/structured_chat/index.d.cts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.cts","names":["StructuredToolInterface","BaseLanguageModelInterface","ToolDefinition","BasePromptTemplate","BaseMessagePromptTemplate","ChatPromptTemplate","AgentStep","Optional","Agent","AgentArgs","AgentRunnableSequence","OutputParserArgs","AgentInput","StructuredChatOutputParserWithRetries","StructuredChatCreatePromptArgs","StructuredChatAgentInput","StructuredChatAgent","Promise","CreateStructuredChatAgentParams","createStructuredChatAgent","llm","tools","prompt","streamRunnable","_langchain_core_agents0","AgentAction","AgentFinish"],"sources":["../../../src/agents/structured_chat/index.d.ts"],"sourcesContent":["import type { StructuredToolInterface } from \"@langchain/core/tools\";\nimport { type BaseLanguageModelInterface, type ToolDefinition } from \"@langchain/core/language_models/base\";\nimport type { BasePromptTemplate } from \"@langchain/core/prompts\";\nimport { BaseMessagePromptTemplate, ChatPromptTemplate } from \"@langchain/core/prompts\";\nimport { AgentStep } from \"@langchain/core/agents\";\nimport { Optional } from \"../../types/type-utils.js\";\nimport { Agent, AgentArgs, AgentRunnableSequence, OutputParserArgs } from \"../agent.js\";\nimport { AgentInput } from \"../types.js\";\nimport { StructuredChatOutputParserWithRetries } from \"./outputParser.js\";\n/**\n * Interface for arguments used to create a prompt for a\n * StructuredChatAgent.\n */\nexport interface StructuredChatCreatePromptArgs {\n    /** String to put after the list of tools. */\n    suffix?: string;\n    /** String to put before the list of tools. */\n    prefix?: string;\n    /** String to use directly as the human message template. */\n    humanMessageTemplate?: string;\n    /** List of input variables the final prompt will expect. */\n    inputVariables?: string[];\n    /** List of historical prompts from memory.  */\n    memoryPrompts?: BaseMessagePromptTemplate[];\n}\n/**\n * Type for input data for creating a StructuredChatAgent, with the\n * 'outputParser' property made optional.\n */\nexport type StructuredChatAgentInput = Optional<AgentInput, \"outputParser\">;\n/**\n * Agent that interoperates with Structured Tools using React logic.\n * @augments Agent\n */\nexport declare class StructuredChatAgent extends Agent {\n    static lc_name(): string;\n    lc_namespace: string[];\n    constructor(input: StructuredChatAgentInput);\n    _agentType(): \"structured-chat-zero-shot-react-description\";\n    observationPrefix(): string;\n    llmPrefix(): string;\n    _stop(): string[];\n    /**\n     * Validates that all provided tools have a description. Throws an error\n     * if any tool lacks a description.\n     * @param tools Array of StructuredTool instances to validate.\n     */\n    static validateTools(tools: StructuredToolInterface[]): void;\n    /**\n     * Returns a default output parser for the StructuredChatAgent. If an LLM\n     * is provided, it creates an output parser with retry logic from the LLM.\n     * @param fields Optional fields to customize the output parser. Can include an LLM and a list of tool names.\n     * @returns An instance of StructuredChatOutputParserWithRetries.\n     */\n    static getDefaultOutputParser(fields?: OutputParserArgs & {\n        toolNames: string[];\n    }): StructuredChatOutputParserWithRetries;\n    /**\n     * Constructs the agent's scratchpad from a list of steps. If the agent's\n     * scratchpad is not empty, it prepends a message indicating that the\n     * agent has not seen any previous work.\n     * @param steps Array of AgentStep instances to construct the scratchpad from.\n     * @returns A Promise that resolves to a string representing the agent's scratchpad.\n     */\n    constructScratchPad(steps: AgentStep[]): Promise<string>;\n    /**\n     * Creates a string representation of the schemas of the provided tools.\n     * @param tools Array of StructuredTool instances to create the schemas string from.\n     * @returns A string representing the schemas of the provided tools.\n     */\n    static createToolSchemasString(tools: StructuredToolInterface[]): string;\n    /**\n     * Create prompt in the style of the agent.\n     *\n     * @param tools - List of tools the agent will have access to, used to format the prompt.\n     * @param args - Arguments to create the prompt with.\n     * @param args.suffix - String to put after the list of tools.\n     * @param args.prefix - String to put before the list of tools.\n     * @param args.inputVariables List of input variables the final prompt will expect.\n     * @param args.memoryPrompts List of historical prompts from memory.\n     */\n    static createPrompt(tools: StructuredToolInterface[], args?: StructuredChatCreatePromptArgs): ChatPromptTemplate<any, any>;\n    /**\n     * Creates a StructuredChatAgent from an LLM and a list of tools.\n     * Validates the tools, creates a prompt, and sets up an LLM chain for the\n     * agent.\n     * @param llm BaseLanguageModel instance to create the agent from.\n     * @param tools Array of StructuredTool instances to create the agent from.\n     * @param args Optional arguments to customize the creation of the agent. Can include arguments for creating the prompt and AgentArgs.\n     * @returns A new instance of StructuredChatAgent.\n     */\n    static fromLLMAndTools(llm: BaseLanguageModelInterface, tools: StructuredToolInterface[], args?: StructuredChatCreatePromptArgs & AgentArgs): StructuredChatAgent;\n}\n/**\n * Params used by the createStructuredChatAgent function.\n */\nexport type CreateStructuredChatAgentParams = {\n    /** LLM to use as the agent. */\n    llm: BaseLanguageModelInterface;\n    /** Tools this agent has access to. */\n    tools: (StructuredToolInterface | ToolDefinition)[];\n    /**\n     * The prompt to use. Must have input keys for\n     * `tools`, `tool_names`, and `agent_scratchpad`.\n     */\n    prompt: BasePromptTemplate;\n    /**\n     * Whether to invoke the underlying model in streaming mode,\n     * allowing streaming of intermediate steps. Defaults to true.\n     */\n    streamRunnable?: boolean;\n};\n/**\n * Create an agent aimed at supporting tools with multiple inputs.\n * @param params Params required to create the agent. Includes an LLM, tools, and prompt.\n * @returns A runnable sequence representing an agent. It takes as input all the same input\n *     variables as the prompt passed in does. It returns as output either an\n *     AgentAction or AgentFinish.\n *\n * @example\n * ```typescript\n * import { AgentExecutor, createStructuredChatAgent } from \"langchain/agents\";\n * import { pull } from \"langchain/hub\";\n * import type { ChatPromptTemplate } from \"@langchain/core/prompts\";\n * import { AIMessage, HumanMessage } from \"@langchain/core/messages\";\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n *\n * // Define the tools the agent will have access to.\n * const tools = [...];\n *\n * // Get the prompt to use - you can modify this!\n * // If you want to see the prompt in full, you can at:\n * // https://smith.langchain.com/hub/hwchase17/structured-chat-agent\n * const prompt = await pull<ChatPromptTemplate>(\n *   \"hwchase17/structured-chat-agent\"\n * );\n *\n * const llm = new ChatOpenAI({\n *   temperature: 0,\n *   model: \"gpt-3.5-turbo-1106\",\n * });\n *\n * const agent = await createStructuredChatAgent({\n *   llm,\n *   tools,\n *   prompt,\n * });\n *\n * const agentExecutor = new AgentExecutor({\n *   agent,\n *   tools,\n * });\n *\n * const result = await agentExecutor.invoke({\n *   input: \"what is LangChain?\",\n * });\n *\n * // With chat history\n * const result2 = await agentExecutor.invoke({\n *   input: \"what's my name?\",\n *   chat_history: [\n *     new HumanMessage(\"hi! my name is cob\"),\n *     new AIMessage(\"Hello Cob! How can I assist you today?\"),\n *   ],\n * });\n * ```\n */\nexport declare function createStructuredChatAgent({ llm, tools, prompt, streamRunnable }: CreateStructuredChatAgentParams): Promise<AgentRunnableSequence<{\n    steps: AgentStep[];\n}, import(\"@langchain/core/agents\").AgentAction | import(\"@langchain/core/agents\").AgentFinish>>;\n//# sourceMappingURL=index.d.ts.map"],"mappings":";;;;;;;;;;;;;;AAaA;AAgBA;AAKqBgB,UArBJF,8BAAAA,CAqBuB;EAGjBC;EAUSf,MAAAA,CAAAA,EAAAA,MAAAA;EAOWW;EAEnCE,MAAAA,CAAAA,EAAAA,MAAAA;EAQuBP;EAAcW,oBAAAA,CAAAA,EAAAA,MAAAA;EAMHjB;EAWXA,cAAAA,CAAAA,EAAAA,MAAAA,EAAAA;EAAkCc;EAAiCT,aAAAA,CAAAA,EA1D9ED,yBA0D8EC,EAAAA;;;;;;AA/CjDG,KALrCO,wBAAAA,GAA2BR,QAKUC,CALDI,UAKCJ,EAAAA,cAAAA,CAAAA;AAAK;AA8DtD;;;AAIsCN,cAlEjBc,mBAAAA,SAA4BR,KAAAA,CAkEXN;EAK1BC,OAAAA,OAAAA,CAAAA,CAAAA,EAAAA,MAAAA;EAAkB,YAAA,EAAA,MAAA,EAAA;EA+DNgB,WAAAA,CAAAA,KAAAA,EAnIDJ,wBAmI0B;EAAGK,UAAAA,CAAAA,CAAAA,EAAAA,6CAAAA;EAAKC,iBAAAA,CAAAA,CAAAA,EAAAA,MAAAA;EAAOC,SAAAA,CAAAA,CAAAA,EAAAA,MAAAA;EAAQC,KAAAA,CAAAA,CAAAA,EAAAA,MAAAA,EAAAA;EAAkBL;;;;;EAAkCD,OAAAA,aAAAA,CAAAA,KAAAA,EAzH5FjB,uBAyH4FiB,EAAAA,CAAAA,EAAAA,IAAAA;EAAO;;;;;;yCAlHxFN;;MAEnCE;;;;;;;;6BAQuBP,cAAcW;;;;;;wCAMHjB;;;;;;;;;;;6BAWXA,kCAAkCc,iCAAiCT;;;;;;;;;;8BAUlEJ,mCAAmCD,kCAAkCc,iCAAiCL,YAAYO;;;;;KAKtIE,+BAAAA;;OAEHjB;;UAEGD,0BAA0BE;;;;;UAK1BC;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBA+DYgB,yBAAAA;;;;;GAAkED,kCAAkCD,QAAQP;SACzHJ;GAASkB,uBAAAA,CACgBC,WAAAA,GAAWD,uBAAAA,CAAoCE"}
+{"version":3,"file":"index.d.cts","names":[],"sources":["../../../src/agents/structured_chat/index.ts"],"mappings":";;;;;;;;;;;;;AAyCA;;UAAiB,8BAAA;EAU0B;EARzC,MAAA;EAEA;EAAA,MAAA;EAIA;EAFA,oBAAA;EAIgB;EAFhB,cAAA;EAEyC;EAAzC,aAAA,GAAgB,yBAAA;AAAA;;;;AAalB;KANY,wBAAA,GAA2B,QAAA,CAAS,UAAA;;;;;cAMnC,mBAAA,SAA4B,KAAA;EAAA,OAChC,OAAA,CAAA;EAIP,YAAA;EAEA,WAAA,CAAY,KAAA,EAAO,wBAAA;EAMnB,UAAA,CAAA;EAIA,iBAAA,CAAA;EAIA,SAAA,CAAA;EAIA,KAAA,CAAA;EAkIS;;;;;EAAA,OAzHF,aAAA,CAAc,KAAA,EAAO,uBAAA;EAlCgB;;;;;;EAAA,OAkDrC,sBAAA,CACL,MAAA,GAAS,gBAAA;IACP,SAAA;EAAA,IACD,qCAAA;EAhCH;;;;;;;EAmDM,mBAAA,CAAoB,KAAA,EAAO,SAAA,KAAc,OAAA;EArB7C;;;;;EAAA,OAkCK,uBAAA,CAAwB,KAAA,EAAO,uBAAA;EAA/B;;;;;;;;;;EAAA,OAyBA,YAAA,CACL,KAAA,EAAO,uBAAA,IACP,IAAA,GAAO,8BAAA,GAA8B,kBAAA;EA0CrC;;;;;;;;;EAAA,OADK,eAAA,CACL,GAAA,EAAK,0BAAA,EACL,KAAA,EAAO,uBAAA,IACP,IAAA,GAAO,8BAAA,GAAiC,SAAA,GAAS,mBAAA;AAAA;;;;KA2BzC,+BAAA;EASF,+BAPR,GAAA,EAAK,0BAAA,EAOqB;EAL1B,KAAA,GAAQ,uBAAA,GAA0B,cAAA;EAF7B;;;;EAOL,MAAA,EAAQ,kBAAA;EAAA;;;;EAKR,cAAA;AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBA2DoB,yBAAA,CAAA;EACpB,GAAA;EACA,KAAA;EACA,MAAA;EACA;AAAA,GACC,+BAAA,GAA+B,OAAA,CAAA,qBAAA"}

package/dist/agents/structured_chat/index.d.ts

@@ -5,11 +5,10 @@ import { StructuredChatOutputParserWithRetries } from "./outputParser.js";
 import { BaseMessagePromptTemplate, BasePromptTemplate, ChatPromptTemplate } from "@langchain/core/prompts";
 import { StructuredToolInterface } from "@langchain/core/tools";
 import { BaseLanguageModelInterface, ToolDefinition } from "@langchain/core/language_models/base";
-import * as _langchain_core_agents5 from "@langchain/core/agents";
+import * as _langchain_core_agents0 from "@langchain/core/agents";
 import { AgentStep } from "@langchain/core/agents";
 
 //#region src/agents/structured_chat/index.d.ts
-
 /**
  * Interface for arguments used to create a prompt for a
  * StructuredChatAgent.
@@ -98,9 +97,7 @@ declare class StructuredChatAgent extends Agent {
  * Params used by the createStructuredChatAgent function.
  */
 type CreateStructuredChatAgentParams = {
-  /** LLM to use as the agent. */
-  llm: BaseLanguageModelInterface;
-  /** Tools this agent has access to. */
+  /** LLM to use as the agent. */llm: BaseLanguageModelInterface; /** Tools this agent has access to. */
   tools: (StructuredToolInterface | ToolDefinition)[];
   /**
    * The prompt to use. Must have input keys for
@@ -176,7 +173,7 @@ declare function createStructuredChatAgent({
   streamRunnable
 }: CreateStructuredChatAgentParams): Promise<AgentRunnableSequence<{
   steps: AgentStep[];
-}, _langchain_core_agents5.AgentAction | _langchain_core_agents5.AgentFinish>>;
+}, _langchain_core_agents0.AgentAction | _langchain_core_agents0.AgentFinish>>;
 //#endregion
 export { CreateStructuredChatAgentParams, StructuredChatAgent, StructuredChatAgentInput, StructuredChatCreatePromptArgs, createStructuredChatAgent };
 //# sourceMappingURL=index.d.ts.map

package/dist/agents/structured_chat/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","names":["StructuredToolInterface","BaseLanguageModelInterface","ToolDefinition","BasePromptTemplate","BaseMessagePromptTemplate","ChatPromptTemplate","AgentStep","Optional","Agent","AgentArgs","AgentRunnableSequence","OutputParserArgs","AgentInput","StructuredChatOutputParserWithRetries","StructuredChatCreatePromptArgs","StructuredChatAgentInput","StructuredChatAgent","Promise","CreateStructuredChatAgentParams","createStructuredChatAgent","llm","tools","prompt","streamRunnable","_langchain_core_agents5","AgentAction","AgentFinish"],"sources":["../../../src/agents/structured_chat/index.d.ts"],"sourcesContent":["import type { StructuredToolInterface } from \"@langchain/core/tools\";\nimport { type BaseLanguageModelInterface, type ToolDefinition } from \"@langchain/core/language_models/base\";\nimport type { BasePromptTemplate } from \"@langchain/core/prompts\";\nimport { BaseMessagePromptTemplate, ChatPromptTemplate } from \"@langchain/core/prompts\";\nimport { AgentStep } from \"@langchain/core/agents\";\nimport { Optional } from \"../../types/type-utils.js\";\nimport { Agent, AgentArgs, AgentRunnableSequence, OutputParserArgs } from \"../agent.js\";\nimport { AgentInput } from \"../types.js\";\nimport { StructuredChatOutputParserWithRetries } from \"./outputParser.js\";\n/**\n * Interface for arguments used to create a prompt for a\n * StructuredChatAgent.\n */\nexport interface StructuredChatCreatePromptArgs {\n    /** String to put after the list of tools. */\n    suffix?: string;\n    /** String to put before the list of tools. */\n    prefix?: string;\n    /** String to use directly as the human message template. */\n    humanMessageTemplate?: string;\n    /** List of input variables the final prompt will expect. */\n    inputVariables?: string[];\n    /** List of historical prompts from memory.  */\n    memoryPrompts?: BaseMessagePromptTemplate[];\n}\n/**\n * Type for input data for creating a StructuredChatAgent, with the\n * 'outputParser' property made optional.\n */\nexport type StructuredChatAgentInput = Optional<AgentInput, \"outputParser\">;\n/**\n * Agent that interoperates with Structured Tools using React logic.\n * @augments Agent\n */\nexport declare class StructuredChatAgent extends Agent {\n    static lc_name(): string;\n    lc_namespace: string[];\n    constructor(input: StructuredChatAgentInput);\n    _agentType(): \"structured-chat-zero-shot-react-description\";\n    observationPrefix(): string;\n    llmPrefix(): string;\n    _stop(): string[];\n    /**\n     * Validates that all provided tools have a description. Throws an error\n     * if any tool lacks a description.\n     * @param tools Array of StructuredTool instances to validate.\n     */\n    static validateTools(tools: StructuredToolInterface[]): void;\n    /**\n     * Returns a default output parser for the StructuredChatAgent. If an LLM\n     * is provided, it creates an output parser with retry logic from the LLM.\n     * @param fields Optional fields to customize the output parser. Can include an LLM and a list of tool names.\n     * @returns An instance of StructuredChatOutputParserWithRetries.\n     */\n    static getDefaultOutputParser(fields?: OutputParserArgs & {\n        toolNames: string[];\n    }): StructuredChatOutputParserWithRetries;\n    /**\n     * Constructs the agent's scratchpad from a list of steps. If the agent's\n     * scratchpad is not empty, it prepends a message indicating that the\n     * agent has not seen any previous work.\n     * @param steps Array of AgentStep instances to construct the scratchpad from.\n     * @returns A Promise that resolves to a string representing the agent's scratchpad.\n     */\n    constructScratchPad(steps: AgentStep[]): Promise<string>;\n    /**\n     * Creates a string representation of the schemas of the provided tools.\n     * @param tools Array of StructuredTool instances to create the schemas string from.\n     * @returns A string representing the schemas of the provided tools.\n     */\n    static createToolSchemasString(tools: StructuredToolInterface[]): string;\n    /**\n     * Create prompt in the style of the agent.\n     *\n     * @param tools - List of tools the agent will have access to, used to format the prompt.\n     * @param args - Arguments to create the prompt with.\n     * @param args.suffix - String to put after the list of tools.\n     * @param args.prefix - String to put before the list of tools.\n     * @param args.inputVariables List of input variables the final prompt will expect.\n     * @param args.memoryPrompts List of historical prompts from memory.\n     */\n    static createPrompt(tools: StructuredToolInterface[], args?: StructuredChatCreatePromptArgs): ChatPromptTemplate<any, any>;\n    /**\n     * Creates a StructuredChatAgent from an LLM and a list of tools.\n     * Validates the tools, creates a prompt, and sets up an LLM chain for the\n     * agent.\n     * @param llm BaseLanguageModel instance to create the agent from.\n     * @param tools Array of StructuredTool instances to create the agent from.\n     * @param args Optional arguments to customize the creation of the agent. Can include arguments for creating the prompt and AgentArgs.\n     * @returns A new instance of StructuredChatAgent.\n     */\n    static fromLLMAndTools(llm: BaseLanguageModelInterface, tools: StructuredToolInterface[], args?: StructuredChatCreatePromptArgs & AgentArgs): StructuredChatAgent;\n}\n/**\n * Params used by the createStructuredChatAgent function.\n */\nexport type CreateStructuredChatAgentParams = {\n    /** LLM to use as the agent. */\n    llm: BaseLanguageModelInterface;\n    /** Tools this agent has access to. */\n    tools: (StructuredToolInterface | ToolDefinition)[];\n    /**\n     * The prompt to use. Must have input keys for\n     * `tools`, `tool_names`, and `agent_scratchpad`.\n     */\n    prompt: BasePromptTemplate;\n    /**\n     * Whether to invoke the underlying model in streaming mode,\n     * allowing streaming of intermediate steps. Defaults to true.\n     */\n    streamRunnable?: boolean;\n};\n/**\n * Create an agent aimed at supporting tools with multiple inputs.\n * @param params Params required to create the agent. Includes an LLM, tools, and prompt.\n * @returns A runnable sequence representing an agent. It takes as input all the same input\n *     variables as the prompt passed in does. It returns as output either an\n *     AgentAction or AgentFinish.\n *\n * @example\n * ```typescript\n * import { AgentExecutor, createStructuredChatAgent } from \"langchain/agents\";\n * import { pull } from \"langchain/hub\";\n * import type { ChatPromptTemplate } from \"@langchain/core/prompts\";\n * import { AIMessage, HumanMessage } from \"@langchain/core/messages\";\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n *\n * // Define the tools the agent will have access to.\n * const tools = [...];\n *\n * // Get the prompt to use - you can modify this!\n * // If you want to see the prompt in full, you can at:\n * // https://smith.langchain.com/hub/hwchase17/structured-chat-agent\n * const prompt = await pull<ChatPromptTemplate>(\n *   \"hwchase17/structured-chat-agent\"\n * );\n *\n * const llm = new ChatOpenAI({\n *   temperature: 0,\n *   model: \"gpt-3.5-turbo-1106\",\n * });\n *\n * const agent = await createStructuredChatAgent({\n *   llm,\n *   tools,\n *   prompt,\n * });\n *\n * const agentExecutor = new AgentExecutor({\n *   agent,\n *   tools,\n * });\n *\n * const result = await agentExecutor.invoke({\n *   input: \"what is LangChain?\",\n * });\n *\n * // With chat history\n * const result2 = await agentExecutor.invoke({\n *   input: \"what's my name?\",\n *   chat_history: [\n *     new HumanMessage(\"hi! my name is cob\"),\n *     new AIMessage(\"Hello Cob! How can I assist you today?\"),\n *   ],\n * });\n * ```\n */\nexport declare function createStructuredChatAgent({ llm, tools, prompt, streamRunnable }: CreateStructuredChatAgentParams): Promise<AgentRunnableSequence<{\n    steps: AgentStep[];\n}, import(\"@langchain/core/agents\").AgentAction | import(\"@langchain/core/agents\").AgentFinish>>;\n//# sourceMappingURL=index.d.ts.map"],"mappings":";;;;;;;;;;;;;;AAaA;AAgBA;AAKqBgB,UArBJF,8BAAAA,CAqBuB;EAGjBC;EAUSf,MAAAA,CAAAA,EAAAA,MAAAA;EAOWW;EAEnCE,MAAAA,CAAAA,EAAAA,MAAAA;EAQuBP;EAAcW,oBAAAA,CAAAA,EAAAA,MAAAA;EAMHjB;EAWXA,cAAAA,CAAAA,EAAAA,MAAAA,EAAAA;EAAkCc;EAAiCT,aAAAA,CAAAA,EA1D9ED,yBA0D8EC,EAAAA;;;;;;AA/CjDG,KALrCO,wBAAAA,GAA2BR,QAKUC,CALDI,UAKCJ,EAAAA,cAAAA,CAAAA;AAAK;AA8DtD;;;AAIsCN,cAlEjBc,mBAAAA,SAA4BR,KAAAA,CAkEXN;EAK1BC,OAAAA,OAAAA,CAAAA,CAAAA,EAAAA,MAAAA;EAAkB,YAAA,EAAA,MAAA,EAAA;EA+DNgB,WAAAA,CAAAA,KAAAA,EAnIDJ,wBAmI0B;EAAGK,UAAAA,CAAAA,CAAAA,EAAAA,6CAAAA;EAAKC,iBAAAA,CAAAA,CAAAA,EAAAA,MAAAA;EAAOC,SAAAA,CAAAA,CAAAA,EAAAA,MAAAA;EAAQC,KAAAA,CAAAA,CAAAA,EAAAA,MAAAA,EAAAA;EAAkBL;;;;;EAAkCD,OAAAA,aAAAA,CAAAA,KAAAA,EAzH5FjB,uBAyH4FiB,EAAAA,CAAAA,EAAAA,IAAAA;EAAO;;;;;;yCAlHxFN;;MAEnCE;;;;;;;;6BAQuBP,cAAcW;;;;;;wCAMHjB;;;;;;;;;;;6BAWXA,kCAAkCc,iCAAiCT;;;;;;;;;;8BAUlEJ,mCAAmCD,kCAAkCc,iCAAiCL,YAAYO;;;;;KAKtIE,+BAAAA;;OAEHjB;;UAEGD,0BAA0BE;;;;;UAK1BC;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBA+DYgB,yBAAAA;;;;;GAAkED,kCAAkCD,QAAQP;SACzHJ;GAASkB,uBAAAA,CACgBC,WAAAA,GAAWD,uBAAAA,CAAoCE"}
+{"version":3,"file":"index.d.ts","names":[],"sources":["../../../src/agents/structured_chat/index.ts"],"mappings":";;;;;;;;;;;;;AAyCA;;UAAiB,8BAAA;EAU0B;EARzC,MAAA;EAEA;EAAA,MAAA;EAIA;EAFA,oBAAA;EAIgB;EAFhB,cAAA;EAEyC;EAAzC,aAAA,GAAgB,yBAAA;AAAA;;;;AAalB;KANY,wBAAA,GAA2B,QAAA,CAAS,UAAA;;;;;cAMnC,mBAAA,SAA4B,KAAA;EAAA,OAChC,OAAA,CAAA;EAIP,YAAA;EAEA,WAAA,CAAY,KAAA,EAAO,wBAAA;EAMnB,UAAA,CAAA;EAIA,iBAAA,CAAA;EAIA,SAAA,CAAA;EAIA,KAAA,CAAA;EAkIS;;;;;EAAA,OAzHF,aAAA,CAAc,KAAA,EAAO,uBAAA;EAlCgB;;;;;;EAAA,OAkDrC,sBAAA,CACL,MAAA,GAAS,gBAAA;IACP,SAAA;EAAA,IACD,qCAAA;EAhCH;;;;;;;EAmDM,mBAAA,CAAoB,KAAA,EAAO,SAAA,KAAc,OAAA;EArB7C;;;;;EAAA,OAkCK,uBAAA,CAAwB,KAAA,EAAO,uBAAA;EAA/B;;;;;;;;;;EAAA,OAyBA,YAAA,CACL,KAAA,EAAO,uBAAA,IACP,IAAA,GAAO,8BAAA,GAA8B,kBAAA;EA0CrC;;;;;;;;;EAAA,OADK,eAAA,CACL,GAAA,EAAK,0BAAA,EACL,KAAA,EAAO,uBAAA,IACP,IAAA,GAAO,8BAAA,GAAiC,SAAA,GAAS,mBAAA;AAAA;;;;KA2BzC,+BAAA;EASF,+BAPR,GAAA,EAAK,0BAAA,EAOqB;EAL1B,KAAA,GAAQ,uBAAA,GAA0B,cAAA;EAF7B;;;;EAOL,MAAA,EAAQ,kBAAA;EAAA;;;;EAKR,cAAA;AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;iBA2DoB,yBAAA,CAAA;EACpB,GAAA;EACA,KAAA;EACA,MAAA;EACA;AAAA,GACC,+BAAA,GAA+B,OAAA,CAAA,qBAAA"}

package/dist/agents/structured_chat/index.js

@@ -1,9 +1,9 @@
+import { formatLogToString } from "../format_scratchpad/log.js";
 import { Agent, AgentRunnableSequence } from "../agent.js";
 import { LLMChain } from "../../chains/llm_chain.js";
 import { FORMAT_INSTRUCTIONS, PREFIX, SUFFIX } from "./prompt.js";
 import { StructuredChatOutputParserWithRetries } from "./outputParser.js";
 import { renderTextDescriptionAndArgs } from "../../tools/render.js";
-import { formatLogToString } from "../format_scratchpad/log.js";
 import { ChatPromptTemplate, HumanMessagePromptTemplate, PromptTemplate, SystemMessagePromptTemplate } from "@langchain/core/prompts";
 import { RunnablePassthrough } from "@langchain/core/runnables";
 import { isOpenAITool } from "@langchain/core/language_models/base";
@@ -101,14 +101,13 @@ var StructuredChatAgent = class StructuredChatAgent extends Agent {
 	*/
 	static createPrompt(tools, args) {
 		const { prefix = PREFIX, suffix = SUFFIX, inputVariables = ["input", "agent_scratchpad"], humanMessageTemplate = "{input}\n\n{agent_scratchpad}", memoryPrompts = [] } = args ?? {};
-		const template = [
-			prefix,
-			FORMAT_INSTRUCTIONS,
-			suffix
-		].join("\n\n");
 		const messages = [
 			new SystemMessagePromptTemplate(new PromptTemplate({
-				template,
+				template: [
+					prefix,
+					FORMAT_INSTRUCTIONS,
+					suffix
+				].join("\n\n"),
 				inputVariables,
 				partialVariables: {
 					tool_schemas: StructuredChatAgent.createToolSchemasString(tools),
@@ -139,13 +138,12 @@ var StructuredChatAgent = class StructuredChatAgent extends Agent {
 			llm,
 			toolNames: tools.map((tool) => tool.name)
 		});
-		const chain = new LLMChain({
-			prompt,
-			llm,
-			callbacks: args?.callbacks
-		});
 		return new StructuredChatAgent({
-			llmChain: chain,
+			llmChain: new LLMChain({
+				prompt,
+				llm,
+				callbacks: args?.callbacks
+			}),
 			outputParser,
 			allowedTools: tools.map((t) => t.name)
 		});
@@ -223,7 +221,7 @@ async function createStructuredChatAgent({ llm, tools, prompt, streamRunnable })
 		tool_names: toolNames.join(", ")
 	});
 	const llmWithStop = llm.withConfig({ stop: ["Observation"] });
-	const agent = AgentRunnableSequence.fromRunnables([
+	return AgentRunnableSequence.fromRunnables([
 		RunnablePassthrough.assign({ agent_scratchpad: (input) => formatLogToString(input.steps) }),
 		partialedPrompt,
 		llmWithStop,
@@ -233,7 +231,6 @@ async function createStructuredChatAgent({ llm, tools, prompt, streamRunnable })
 		streamRunnable,
 		singleAction: true
 	});
-	return agent;
 }
 
 //#endregion

package/dist/agents/structured_chat/index.js.map

@@ -1 +1 @@
-{"version":3,"file":"index.js","names":["input: StructuredChatAgentInput","tools: StructuredToolInterface[]","fields?: OutputParserArgs & {\n      toolNames: string[];\n    }","steps: AgentStep[]","args?: StructuredChatCreatePromptArgs","llm: BaseLanguageModelInterface","args?: StructuredChatCreatePromptArgs & AgentArgs","toolNames: string[]","input: { steps: AgentStep[] }"],"sources":["../../../src/agents/structured_chat/index.ts"],"sourcesContent":["import type { StructuredToolInterface } from \"@langchain/core/tools\";\nimport {\n  isOpenAITool,\n  type BaseLanguageModel,\n  type BaseLanguageModelInterface,\n  type ToolDefinition,\n} from \"@langchain/core/language_models/base\";\nimport { RunnablePassthrough } from \"@langchain/core/runnables\";\nimport type { BasePromptTemplate } from \"@langchain/core/prompts\";\nimport {\n  BaseMessagePromptTemplate,\n  ChatPromptTemplate,\n  HumanMessagePromptTemplate,\n  SystemMessagePromptTemplate,\n  PromptTemplate,\n} from \"@langchain/core/prompts\";\nimport { AgentStep } from \"@langchain/core/agents\";\nimport { isStructuredTool } from \"@langchain/core/utils/function_calling\";\nimport { isInteropZodSchema } from \"@langchain/core/utils/types\";\nimport {\n  type JsonSchema7Type,\n  toJsonSchema,\n} from \"@langchain/core/utils/json_schema\";\nimport { LLMChain } from \"../../chains/llm_chain.js\";\nimport { Optional } from \"../../types/type-utils.js\";\nimport {\n  Agent,\n  AgentArgs,\n  AgentRunnableSequence,\n  OutputParserArgs,\n} from \"../agent.js\";\nimport { AgentInput } from \"../types.js\";\nimport { StructuredChatOutputParserWithRetries } from \"./outputParser.js\";\nimport { FORMAT_INSTRUCTIONS, PREFIX, SUFFIX } from \"./prompt.js\";\nimport { renderTextDescriptionAndArgs } from \"../../tools/render.js\";\nimport { formatLogToString } from \"../format_scratchpad/log.js\";\n\n/**\n * Interface for arguments used to create a prompt for a\n * StructuredChatAgent.\n */\nexport interface StructuredChatCreatePromptArgs {\n  /** String to put after the list of tools. */\n  suffix?: string;\n  /** String to put before the list of tools. */\n  prefix?: string;\n  /** String to use directly as the human message template. */\n  humanMessageTemplate?: string;\n  /** List of input variables the final prompt will expect. */\n  inputVariables?: string[];\n  /** List of historical prompts from memory.  */\n  memoryPrompts?: BaseMessagePromptTemplate[];\n}\n\n/**\n * Type for input data for creating a StructuredChatAgent, with the\n * 'outputParser' property made optional.\n */\nexport type StructuredChatAgentInput = Optional<AgentInput, \"outputParser\">;\n\n/**\n * Agent that interoperates with Structured Tools using React logic.\n * @augments Agent\n */\nexport class StructuredChatAgent extends Agent {\n  static lc_name() {\n    return \"StructuredChatAgent\";\n  }\n\n  lc_namespace = [\"langchain\", \"agents\", \"structured_chat\"];\n\n  constructor(input: StructuredChatAgentInput) {\n    const outputParser =\n      input?.outputParser ?? StructuredChatAgent.getDefaultOutputParser();\n    super({ ...input, outputParser });\n  }\n\n  _agentType() {\n    return \"structured-chat-zero-shot-react-description\" as const;\n  }\n\n  observationPrefix() {\n    return \"Observation: \";\n  }\n\n  llmPrefix() {\n    return \"Thought:\";\n  }\n\n  _stop(): string[] {\n    return [\"Observation:\"];\n  }\n\n  /**\n   * Validates that all provided tools have a description. Throws an error\n   * if any tool lacks a description.\n   * @param tools Array of StructuredTool instances to validate.\n   */\n  static validateTools(tools: StructuredToolInterface[]) {\n    const descriptionlessTool = tools.find((tool) => !tool.description);\n    if (descriptionlessTool) {\n      const msg =\n        `Got a tool ${descriptionlessTool.name} without a description.` +\n        ` This agent requires descriptions for all tools.`;\n      throw new Error(msg);\n    }\n  }\n\n  /**\n   * Returns a default output parser for the StructuredChatAgent. If an LLM\n   * is provided, it creates an output parser with retry logic from the LLM.\n   * @param fields Optional fields to customize the output parser. Can include an LLM and a list of tool names.\n   * @returns An instance of StructuredChatOutputParserWithRetries.\n   */\n  static getDefaultOutputParser(\n    fields?: OutputParserArgs & {\n      toolNames: string[];\n    }\n  ) {\n    if (fields?.llm) {\n      return StructuredChatOutputParserWithRetries.fromLLM(fields.llm, {\n        toolNames: fields.toolNames,\n      });\n    }\n    return new StructuredChatOutputParserWithRetries({\n      toolNames: fields?.toolNames,\n    });\n  }\n\n  /**\n   * Constructs the agent's scratchpad from a list of steps. If the agent's\n   * scratchpad is not empty, it prepends a message indicating that the\n   * agent has not seen any previous work.\n   * @param steps Array of AgentStep instances to construct the scratchpad from.\n   * @returns A Promise that resolves to a string representing the agent's scratchpad.\n   */\n  async constructScratchPad(steps: AgentStep[]): Promise<string> {\n    const agentScratchpad = await super.constructScratchPad(steps);\n    if (agentScratchpad) {\n      return `This was your previous work (but I haven't seen any of it! I only see what you return as final answer):\\n${agentScratchpad}`;\n    }\n    return agentScratchpad;\n  }\n\n  /**\n   * Creates a string representation of the schemas of the provided tools.\n   * @param tools Array of StructuredTool instances to create the schemas string from.\n   * @returns A string representing the schemas of the provided tools.\n   */\n  static createToolSchemasString(tools: StructuredToolInterface[]) {\n    return tools\n      .map((tool) => {\n        const jsonSchema = (\n          isInteropZodSchema(tool.schema)\n            ? toJsonSchema(tool.schema)\n            : tool.schema\n        ) as { properties?: Record<string, JsonSchema7Type> } | undefined;\n        return `${tool.name}: ${tool.description}, args: ${JSON.stringify(\n          jsonSchema?.properties\n        )}`;\n      })\n      .join(\"\\n\");\n  }\n\n  /**\n   * Create prompt in the style of the agent.\n   *\n   * @param tools - List of tools the agent will have access to, used to format the prompt.\n   * @param args - Arguments to create the prompt with.\n   * @param args.suffix - String to put after the list of tools.\n   * @param args.prefix - String to put before the list of tools.\n   * @param args.inputVariables List of input variables the final prompt will expect.\n   * @param args.memoryPrompts List of historical prompts from memory.\n   */\n  static createPrompt(\n    tools: StructuredToolInterface[],\n    args?: StructuredChatCreatePromptArgs\n  ) {\n    const {\n      prefix = PREFIX,\n      suffix = SUFFIX,\n      inputVariables = [\"input\", \"agent_scratchpad\"],\n      humanMessageTemplate = \"{input}\\n\\n{agent_scratchpad}\",\n      memoryPrompts = [],\n    } = args ?? {};\n    const template = [prefix, FORMAT_INSTRUCTIONS, suffix].join(\"\\n\\n\");\n    const messages = [\n      new SystemMessagePromptTemplate(\n        new PromptTemplate({\n          template,\n          inputVariables,\n          partialVariables: {\n            tool_schemas: StructuredChatAgent.createToolSchemasString(tools),\n            tool_names: tools.map((tool) => tool.name).join(\", \"),\n          },\n        })\n      ),\n      ...memoryPrompts,\n      new HumanMessagePromptTemplate(\n        new PromptTemplate({\n          template: humanMessageTemplate,\n          inputVariables,\n        })\n      ),\n    ];\n    return ChatPromptTemplate.fromMessages(messages);\n  }\n\n  /**\n   * Creates a StructuredChatAgent from an LLM and a list of tools.\n   * Validates the tools, creates a prompt, and sets up an LLM chain for the\n   * agent.\n   * @param llm BaseLanguageModel instance to create the agent from.\n   * @param tools Array of StructuredTool instances to create the agent from.\n   * @param args Optional arguments to customize the creation of the agent. Can include arguments for creating the prompt and AgentArgs.\n   * @returns A new instance of StructuredChatAgent.\n   */\n  static fromLLMAndTools(\n    llm: BaseLanguageModelInterface,\n    tools: StructuredToolInterface[],\n    args?: StructuredChatCreatePromptArgs & AgentArgs\n  ) {\n    StructuredChatAgent.validateTools(tools);\n    const prompt = StructuredChatAgent.createPrompt(tools, args);\n    const outputParser =\n      args?.outputParser ??\n      StructuredChatAgent.getDefaultOutputParser({\n        llm,\n        toolNames: tools.map((tool) => tool.name),\n      });\n    const chain = new LLMChain({\n      prompt,\n      llm,\n      callbacks: args?.callbacks,\n    });\n\n    return new StructuredChatAgent({\n      llmChain: chain,\n      outputParser,\n      allowedTools: tools.map((t) => t.name),\n    });\n  }\n}\n\n/**\n * Params used by the createStructuredChatAgent function.\n */\nexport type CreateStructuredChatAgentParams = {\n  /** LLM to use as the agent. */\n  llm: BaseLanguageModelInterface;\n  /** Tools this agent has access to. */\n  tools: (StructuredToolInterface | ToolDefinition)[];\n  /**\n   * The prompt to use. Must have input keys for\n   * `tools`, `tool_names`, and `agent_scratchpad`.\n   */\n  prompt: BasePromptTemplate;\n  /**\n   * Whether to invoke the underlying model in streaming mode,\n   * allowing streaming of intermediate steps. Defaults to true.\n   */\n  streamRunnable?: boolean;\n};\n\n/**\n * Create an agent aimed at supporting tools with multiple inputs.\n * @param params Params required to create the agent. Includes an LLM, tools, and prompt.\n * @returns A runnable sequence representing an agent. It takes as input all the same input\n *     variables as the prompt passed in does. It returns as output either an\n *     AgentAction or AgentFinish.\n *\n * @example\n * ```typescript\n * import { AgentExecutor, createStructuredChatAgent } from \"langchain/agents\";\n * import { pull } from \"langchain/hub\";\n * import type { ChatPromptTemplate } from \"@langchain/core/prompts\";\n * import { AIMessage, HumanMessage } from \"@langchain/core/messages\";\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n *\n * // Define the tools the agent will have access to.\n * const tools = [...];\n *\n * // Get the prompt to use - you can modify this!\n * // If you want to see the prompt in full, you can at:\n * // https://smith.langchain.com/hub/hwchase17/structured-chat-agent\n * const prompt = await pull<ChatPromptTemplate>(\n *   \"hwchase17/structured-chat-agent\"\n * );\n *\n * const llm = new ChatOpenAI({\n *   temperature: 0,\n *   model: \"gpt-3.5-turbo-1106\",\n * });\n *\n * const agent = await createStructuredChatAgent({\n *   llm,\n *   tools,\n *   prompt,\n * });\n *\n * const agentExecutor = new AgentExecutor({\n *   agent,\n *   tools,\n * });\n *\n * const result = await agentExecutor.invoke({\n *   input: \"what is LangChain?\",\n * });\n *\n * // With chat history\n * const result2 = await agentExecutor.invoke({\n *   input: \"what's my name?\",\n *   chat_history: [\n *     new HumanMessage(\"hi! my name is cob\"),\n *     new AIMessage(\"Hello Cob! How can I assist you today?\"),\n *   ],\n * });\n * ```\n */\nexport async function createStructuredChatAgent({\n  llm,\n  tools,\n  prompt,\n  streamRunnable,\n}: CreateStructuredChatAgentParams) {\n  const missingVariables = [\"tools\", \"tool_names\", \"agent_scratchpad\"].filter(\n    (v) => !prompt.inputVariables.includes(v)\n  );\n  if (missingVariables.length > 0) {\n    throw new Error(\n      `Provided prompt is missing required input variables: ${JSON.stringify(\n        missingVariables\n      )}`\n    );\n  }\n  let toolNames: string[] = [];\n  if (tools.every(isOpenAITool)) {\n    toolNames = tools.map((tool) => tool.function.name);\n  } else if (tools.every(isStructuredTool)) {\n    toolNames = tools.map((tool) => tool.name);\n  } else {\n    throw new Error(\n      \"All tools must be either OpenAI or Structured tools, not a mix.\"\n    );\n  }\n  const partialedPrompt = await prompt.partial({\n    tools: renderTextDescriptionAndArgs(tools),\n    tool_names: toolNames.join(\", \"),\n  });\n  // TODO: Add .bind to core runnable interface.\n  const llmWithStop = (llm as BaseLanguageModel).withConfig({\n    stop: [\"Observation\"],\n  });\n  const agent = AgentRunnableSequence.fromRunnables(\n    [\n      RunnablePassthrough.assign({\n        agent_scratchpad: (input: { steps: AgentStep[] }) =>\n          formatLogToString(input.steps),\n      }),\n      partialedPrompt,\n      llmWithStop,\n      StructuredChatOutputParserWithRetries.fromLLM(llm, {\n        toolNames,\n      }),\n    ],\n    {\n      name: \"StructuredChatAgent\",\n      streamRunnable,\n      singleAction: true,\n    }\n  );\n  return agent;\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;AAgEA,IAAa,sBAAb,MAAa,4BAA4B,MAAM;CAC7C,OAAO,UAAU;AACf,SAAO;CACR;CAED,eAAe;EAAC;EAAa;EAAU;CAAkB;CAEzD,YAAYA,OAAiC;EAC3C,MAAM,eACJ,OAAO,gBAAgB,oBAAoB,wBAAwB;EACrE,MAAM;GAAE,GAAG;GAAO;EAAc,EAAC;CAClC;CAED,aAAa;AACX,SAAO;CACR;CAED,oBAAoB;AAClB,SAAO;CACR;CAED,YAAY;AACV,SAAO;CACR;CAED,QAAkB;AAChB,SAAO,CAAC,cAAe;CACxB;;;;;;CAOD,OAAO,cAAcC,OAAkC;EACrD,MAAM,sBAAsB,MAAM,KAAK,CAAC,SAAS,CAAC,KAAK,YAAY;AACnE,MAAI,qBAAqB;GACvB,MAAM,MACJ,CAAC,WAAW,EAAE,oBAAoB,KAAK,uEAAuB,CACZ;AACpD,SAAM,IAAI,MAAM;EACjB;CACF;;;;;;;CAQD,OAAO,uBACLC,QAGA;AACA,MAAI,QAAQ,IACV,QAAO,sCAAsC,QAAQ,OAAO,KAAK,EAC/D,WAAW,OAAO,UACnB,EAAC;AAEJ,SAAO,IAAI,sCAAsC,EAC/C,WAAW,QAAQ,UACpB;CACF;;;;;;;;CASD,MAAM,oBAAoBC,OAAqC;EAC7D,MAAM,kBAAkB,MAAM,MAAM,oBAAoB,MAAM;AAC9D,MAAI,gBACF,QAAO,CAAC,yGAAyG,EAAE,iBAAiB;AAEtI,SAAO;CACR;;;;;;CAOD,OAAO,wBAAwBF,OAAkC;AAC/D,SAAO,MACJ,IAAI,CAAC,SAAS;GACb,MAAM,aACJ,mBAAmB,KAAK,OAAO,GAC3B,aAAa,KAAK,OAAO,GACzB,KAAK;AAEX,UAAO,GAAG,KAAK,KAAK,EAAE,EAAE,KAAK,YAAY,QAAQ,EAAE,KAAK,UACtD,YAAY,WACb,EAAE;EACJ,EAAC,CACD,KAAK,KAAK;CACd;;;;;;;;;;;CAYD,OAAO,aACLA,OACAG,MACA;EACA,MAAM,EACJ,SAAS,QACT,SAAS,QACT,iBAAiB,CAAC,SAAS,kBAAmB,GAC9C,uBAAuB,iCACvB,gBAAgB,CAAE,GACnB,GAAG,QAAQ,CAAE;EACd,MAAM,WAAW;GAAC;GAAQ;GAAqB;EAAO,EAAC,KAAK,OAAO;EACnE,MAAM,WAAW;GACf,IAAI,4BACF,IAAI,eAAe;IACjB;IACA;IACA,kBAAkB;KAChB,cAAc,oBAAoB,wBAAwB,MAAM;KAChE,YAAY,MAAM,IAAI,CAAC,SAAS,KAAK,KAAK,CAAC,KAAK,KAAK;IACtD;GACF;GAEH,GAAG;GACH,IAAI,2BACF,IAAI,eAAe;IACjB,UAAU;IACV;GACD;EAEJ;AACD,SAAO,mBAAmB,aAAa,SAAS;CACjD;;;;;;;;;;CAWD,OAAO,gBACLC,KACAJ,OACAK,MACA;EACA,oBAAoB,cAAc,MAAM;EACxC,MAAM,SAAS,oBAAoB,aAAa,OAAO,KAAK;EAC5D,MAAM,eACJ,MAAM,gBACN,oBAAoB,uBAAuB;GACzC;GACA,WAAW,MAAM,IAAI,CAAC,SAAS,KAAK,KAAK;EAC1C,EAAC;EACJ,MAAM,QAAQ,IAAI,SAAS;GACzB;GACA;GACA,WAAW,MAAM;EAClB;AAED,SAAO,IAAI,oBAAoB;GAC7B,UAAU;GACV;GACA,cAAc,MAAM,IAAI,CAAC,MAAM,EAAE,KAAK;EACvC;CACF;AACF;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AA8ED,eAAsB,0BAA0B,EAC9C,KACA,OACA,QACA,gBACgC,EAAE;CAClC,MAAM,mBAAmB;EAAC;EAAS;EAAc;CAAmB,EAAC,OACnE,CAAC,MAAM,CAAC,OAAO,eAAe,SAAS,EAAE,CAC1C;AACD,KAAI,iBAAiB,SAAS,EAC5B,OAAM,IAAI,MACR,CAAC,qDAAqD,EAAE,KAAK,UAC3D,iBACD,EAAE;CAGP,IAAIC,YAAsB,CAAE;AAC5B,KAAI,MAAM,MAAM,aAAa,EAC3B,YAAY,MAAM,IAAI,CAAC,SAAS,KAAK,SAAS,KAAK;UAC1C,MAAM,MAAM,iBAAiB,EACtC,YAAY,MAAM,IAAI,CAAC,SAAS,KAAK,KAAK;KAE1C,OAAM,IAAI,MACR;CAGJ,MAAM,kBAAkB,MAAM,OAAO,QAAQ;EAC3C,OAAO,6BAA6B,MAAM;EAC1C,YAAY,UAAU,KAAK,KAAK;CACjC,EAAC;CAEF,MAAM,cAAe,IAA0B,WAAW,EACxD,MAAM,CAAC,aAAc,EACtB,EAAC;CACF,MAAM,QAAQ,sBAAsB,cAClC;EACE,oBAAoB,OAAO,EACzB,kBAAkB,CAACC,UACjB,kBAAkB,MAAM,MAAM,CACjC,EAAC;EACF;EACA;EACA,sCAAsC,QAAQ,KAAK,EACjD,UACD,EAAC;CACH,GACD;EACE,MAAM;EACN;EACA,cAAc;CACf,EACF;AACD,QAAO;AACR"}
+{"version":3,"file":"index.js","names":[],"sources":["../../../src/agents/structured_chat/index.ts"],"sourcesContent":["import type { StructuredToolInterface } from \"@langchain/core/tools\";\nimport {\n  isOpenAITool,\n  type BaseLanguageModel,\n  type BaseLanguageModelInterface,\n  type ToolDefinition,\n} from \"@langchain/core/language_models/base\";\nimport { RunnablePassthrough } from \"@langchain/core/runnables\";\nimport type { BasePromptTemplate } from \"@langchain/core/prompts\";\nimport {\n  BaseMessagePromptTemplate,\n  ChatPromptTemplate,\n  HumanMessagePromptTemplate,\n  SystemMessagePromptTemplate,\n  PromptTemplate,\n} from \"@langchain/core/prompts\";\nimport { AgentStep } from \"@langchain/core/agents\";\nimport { isStructuredTool } from \"@langchain/core/utils/function_calling\";\nimport { isInteropZodSchema } from \"@langchain/core/utils/types\";\nimport {\n  type JsonSchema7Type,\n  toJsonSchema,\n} from \"@langchain/core/utils/json_schema\";\nimport { LLMChain } from \"../../chains/llm_chain.js\";\nimport { Optional } from \"../../types/type-utils.js\";\nimport {\n  Agent,\n  AgentArgs,\n  AgentRunnableSequence,\n  OutputParserArgs,\n} from \"../agent.js\";\nimport { AgentInput } from \"../types.js\";\nimport { StructuredChatOutputParserWithRetries } from \"./outputParser.js\";\nimport { FORMAT_INSTRUCTIONS, PREFIX, SUFFIX } from \"./prompt.js\";\nimport { renderTextDescriptionAndArgs } from \"../../tools/render.js\";\nimport { formatLogToString } from \"../format_scratchpad/log.js\";\n\n/**\n * Interface for arguments used to create a prompt for a\n * StructuredChatAgent.\n */\nexport interface StructuredChatCreatePromptArgs {\n  /** String to put after the list of tools. */\n  suffix?: string;\n  /** String to put before the list of tools. */\n  prefix?: string;\n  /** String to use directly as the human message template. */\n  humanMessageTemplate?: string;\n  /** List of input variables the final prompt will expect. */\n  inputVariables?: string[];\n  /** List of historical prompts from memory.  */\n  memoryPrompts?: BaseMessagePromptTemplate[];\n}\n\n/**\n * Type for input data for creating a StructuredChatAgent, with the\n * 'outputParser' property made optional.\n */\nexport type StructuredChatAgentInput = Optional<AgentInput, \"outputParser\">;\n\n/**\n * Agent that interoperates with Structured Tools using React logic.\n * @augments Agent\n */\nexport class StructuredChatAgent extends Agent {\n  static lc_name() {\n    return \"StructuredChatAgent\";\n  }\n\n  lc_namespace = [\"langchain\", \"agents\", \"structured_chat\"];\n\n  constructor(input: StructuredChatAgentInput) {\n    const outputParser =\n      input?.outputParser ?? StructuredChatAgent.getDefaultOutputParser();\n    super({ ...input, outputParser });\n  }\n\n  _agentType() {\n    return \"structured-chat-zero-shot-react-description\" as const;\n  }\n\n  observationPrefix() {\n    return \"Observation: \";\n  }\n\n  llmPrefix() {\n    return \"Thought:\";\n  }\n\n  _stop(): string[] {\n    return [\"Observation:\"];\n  }\n\n  /**\n   * Validates that all provided tools have a description. Throws an error\n   * if any tool lacks a description.\n   * @param tools Array of StructuredTool instances to validate.\n   */\n  static validateTools(tools: StructuredToolInterface[]) {\n    const descriptionlessTool = tools.find((tool) => !tool.description);\n    if (descriptionlessTool) {\n      const msg =\n        `Got a tool ${descriptionlessTool.name} without a description.` +\n        ` This agent requires descriptions for all tools.`;\n      throw new Error(msg);\n    }\n  }\n\n  /**\n   * Returns a default output parser for the StructuredChatAgent. If an LLM\n   * is provided, it creates an output parser with retry logic from the LLM.\n   * @param fields Optional fields to customize the output parser. Can include an LLM and a list of tool names.\n   * @returns An instance of StructuredChatOutputParserWithRetries.\n   */\n  static getDefaultOutputParser(\n    fields?: OutputParserArgs & {\n      toolNames: string[];\n    }\n  ) {\n    if (fields?.llm) {\n      return StructuredChatOutputParserWithRetries.fromLLM(fields.llm, {\n        toolNames: fields.toolNames,\n      });\n    }\n    return new StructuredChatOutputParserWithRetries({\n      toolNames: fields?.toolNames,\n    });\n  }\n\n  /**\n   * Constructs the agent's scratchpad from a list of steps. If the agent's\n   * scratchpad is not empty, it prepends a message indicating that the\n   * agent has not seen any previous work.\n   * @param steps Array of AgentStep instances to construct the scratchpad from.\n   * @returns A Promise that resolves to a string representing the agent's scratchpad.\n   */\n  async constructScratchPad(steps: AgentStep[]): Promise<string> {\n    const agentScratchpad = await super.constructScratchPad(steps);\n    if (agentScratchpad) {\n      return `This was your previous work (but I haven't seen any of it! I only see what you return as final answer):\\n${agentScratchpad}`;\n    }\n    return agentScratchpad;\n  }\n\n  /**\n   * Creates a string representation of the schemas of the provided tools.\n   * @param tools Array of StructuredTool instances to create the schemas string from.\n   * @returns A string representing the schemas of the provided tools.\n   */\n  static createToolSchemasString(tools: StructuredToolInterface[]) {\n    return tools\n      .map((tool) => {\n        const jsonSchema = (\n          isInteropZodSchema(tool.schema)\n            ? toJsonSchema(tool.schema)\n            : tool.schema\n        ) as { properties?: Record<string, JsonSchema7Type> } | undefined;\n        return `${tool.name}: ${tool.description}, args: ${JSON.stringify(\n          jsonSchema?.properties\n        )}`;\n      })\n      .join(\"\\n\");\n  }\n\n  /**\n   * Create prompt in the style of the agent.\n   *\n   * @param tools - List of tools the agent will have access to, used to format the prompt.\n   * @param args - Arguments to create the prompt with.\n   * @param args.suffix - String to put after the list of tools.\n   * @param args.prefix - String to put before the list of tools.\n   * @param args.inputVariables List of input variables the final prompt will expect.\n   * @param args.memoryPrompts List of historical prompts from memory.\n   */\n  static createPrompt(\n    tools: StructuredToolInterface[],\n    args?: StructuredChatCreatePromptArgs\n  ) {\n    const {\n      prefix = PREFIX,\n      suffix = SUFFIX,\n      inputVariables = [\"input\", \"agent_scratchpad\"],\n      humanMessageTemplate = \"{input}\\n\\n{agent_scratchpad}\",\n      memoryPrompts = [],\n    } = args ?? {};\n    const template = [prefix, FORMAT_INSTRUCTIONS, suffix].join(\"\\n\\n\");\n    const messages = [\n      new SystemMessagePromptTemplate(\n        new PromptTemplate({\n          template,\n          inputVariables,\n          partialVariables: {\n            tool_schemas: StructuredChatAgent.createToolSchemasString(tools),\n            tool_names: tools.map((tool) => tool.name).join(\", \"),\n          },\n        })\n      ),\n      ...memoryPrompts,\n      new HumanMessagePromptTemplate(\n        new PromptTemplate({\n          template: humanMessageTemplate,\n          inputVariables,\n        })\n      ),\n    ];\n    return ChatPromptTemplate.fromMessages(messages);\n  }\n\n  /**\n   * Creates a StructuredChatAgent from an LLM and a list of tools.\n   * Validates the tools, creates a prompt, and sets up an LLM chain for the\n   * agent.\n   * @param llm BaseLanguageModel instance to create the agent from.\n   * @param tools Array of StructuredTool instances to create the agent from.\n   * @param args Optional arguments to customize the creation of the agent. Can include arguments for creating the prompt and AgentArgs.\n   * @returns A new instance of StructuredChatAgent.\n   */\n  static fromLLMAndTools(\n    llm: BaseLanguageModelInterface,\n    tools: StructuredToolInterface[],\n    args?: StructuredChatCreatePromptArgs & AgentArgs\n  ) {\n    StructuredChatAgent.validateTools(tools);\n    const prompt = StructuredChatAgent.createPrompt(tools, args);\n    const outputParser =\n      args?.outputParser ??\n      StructuredChatAgent.getDefaultOutputParser({\n        llm,\n        toolNames: tools.map((tool) => tool.name),\n      });\n    const chain = new LLMChain({\n      prompt,\n      llm,\n      callbacks: args?.callbacks,\n    });\n\n    return new StructuredChatAgent({\n      llmChain: chain,\n      outputParser,\n      allowedTools: tools.map((t) => t.name),\n    });\n  }\n}\n\n/**\n * Params used by the createStructuredChatAgent function.\n */\nexport type CreateStructuredChatAgentParams = {\n  /** LLM to use as the agent. */\n  llm: BaseLanguageModelInterface;\n  /** Tools this agent has access to. */\n  tools: (StructuredToolInterface | ToolDefinition)[];\n  /**\n   * The prompt to use. Must have input keys for\n   * `tools`, `tool_names`, and `agent_scratchpad`.\n   */\n  prompt: BasePromptTemplate;\n  /**\n   * Whether to invoke the underlying model in streaming mode,\n   * allowing streaming of intermediate steps. Defaults to true.\n   */\n  streamRunnable?: boolean;\n};\n\n/**\n * Create an agent aimed at supporting tools with multiple inputs.\n * @param params Params required to create the agent. Includes an LLM, tools, and prompt.\n * @returns A runnable sequence representing an agent. It takes as input all the same input\n *     variables as the prompt passed in does. It returns as output either an\n *     AgentAction or AgentFinish.\n *\n * @example\n * ```typescript\n * import { AgentExecutor, createStructuredChatAgent } from \"langchain/agents\";\n * import { pull } from \"langchain/hub\";\n * import type { ChatPromptTemplate } from \"@langchain/core/prompts\";\n * import { AIMessage, HumanMessage } from \"@langchain/core/messages\";\n *\n * import { ChatOpenAI } from \"@langchain/openai\";\n *\n * // Define the tools the agent will have access to.\n * const tools = [...];\n *\n * // Get the prompt to use - you can modify this!\n * // If you want to see the prompt in full, you can at:\n * // https://smith.langchain.com/hub/hwchase17/structured-chat-agent\n * const prompt = await pull<ChatPromptTemplate>(\n *   \"hwchase17/structured-chat-agent\"\n * );\n *\n * const llm = new ChatOpenAI({\n *   temperature: 0,\n *   model: \"gpt-3.5-turbo-1106\",\n * });\n *\n * const agent = await createStructuredChatAgent({\n *   llm,\n *   tools,\n *   prompt,\n * });\n *\n * const agentExecutor = new AgentExecutor({\n *   agent,\n *   tools,\n * });\n *\n * const result = await agentExecutor.invoke({\n *   input: \"what is LangChain?\",\n * });\n *\n * // With chat history\n * const result2 = await agentExecutor.invoke({\n *   input: \"what's my name?\",\n *   chat_history: [\n *     new HumanMessage(\"hi! my name is cob\"),\n *     new AIMessage(\"Hello Cob! How can I assist you today?\"),\n *   ],\n * });\n * ```\n */\nexport async function createStructuredChatAgent({\n  llm,\n  tools,\n  prompt,\n  streamRunnable,\n}: CreateStructuredChatAgentParams) {\n  const missingVariables = [\"tools\", \"tool_names\", \"agent_scratchpad\"].filter(\n    (v) => !prompt.inputVariables.includes(v)\n  );\n  if (missingVariables.length > 0) {\n    throw new Error(\n      `Provided prompt is missing required input variables: ${JSON.stringify(\n        missingVariables\n      )}`\n    );\n  }\n  let toolNames: string[] = [];\n  if (tools.every(isOpenAITool)) {\n    toolNames = tools.map((tool) => tool.function.name);\n  } else if (tools.every(isStructuredTool)) {\n    toolNames = tools.map((tool) => tool.name);\n  } else {\n    throw new Error(\n      \"All tools must be either OpenAI or Structured tools, not a mix.\"\n    );\n  }\n  const partialedPrompt = await prompt.partial({\n    tools: renderTextDescriptionAndArgs(tools),\n    tool_names: toolNames.join(\", \"),\n  });\n  // TODO: Add .bind to core runnable interface.\n  const llmWithStop = (llm as BaseLanguageModel).withConfig({\n    stop: [\"Observation\"],\n  });\n  const agent = AgentRunnableSequence.fromRunnables(\n    [\n      RunnablePassthrough.assign({\n        agent_scratchpad: (input: { steps: AgentStep[] }) =>\n          formatLogToString(input.steps),\n      }),\n      partialedPrompt,\n      llmWithStop,\n      StructuredChatOutputParserWithRetries.fromLLM(llm, {\n        toolNames,\n      }),\n    ],\n    {\n      name: \"StructuredChatAgent\",\n      streamRunnable,\n      singleAction: true,\n    }\n  );\n  return agent;\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;AAgEA,IAAa,sBAAb,MAAa,4BAA4B,MAAM;CAC7C,OAAO,UAAU;AACf,SAAO;;CAGT,eAAe;EAAC;EAAa;EAAU;EAAkB;CAEzD,YAAY,OAAiC;EAC3C,MAAM,eACJ,OAAO,gBAAgB,oBAAoB,wBAAwB;AACrE,QAAM;GAAE,GAAG;GAAO;GAAc,CAAC;;CAGnC,aAAa;AACX,SAAO;;CAGT,oBAAoB;AAClB,SAAO;;CAGT,YAAY;AACV,SAAO;;CAGT,QAAkB;AAChB,SAAO,CAAC,eAAe;;;;;;;CAQzB,OAAO,cAAc,OAAkC;EACrD,MAAM,sBAAsB,MAAM,MAAM,SAAS,CAAC,KAAK,YAAY;AACnE,MAAI,qBAAqB;GACvB,MAAM,MACJ,cAAc,oBAAoB,KAAK;AAEzC,SAAM,IAAI,MAAM,IAAI;;;;;;;;;CAUxB,OAAO,uBACL,QAGA;AACA,MAAI,QAAQ,IACV,QAAO,sCAAsC,QAAQ,OAAO,KAAK,EAC/D,WAAW,OAAO,WACnB,CAAC;AAEJ,SAAO,IAAI,sCAAsC,EAC/C,WAAW,QAAQ,WACpB,CAAC;;;;;;;;;CAUJ,MAAM,oBAAoB,OAAqC;EAC7D,MAAM,kBAAkB,MAAM,MAAM,oBAAoB,MAAM;AAC9D,MAAI,gBACF,QAAO,4GAA4G;AAErH,SAAO;;;;;;;CAQT,OAAO,wBAAwB,OAAkC;AAC/D,SAAO,MACJ,KAAK,SAAS;GACb,MAAM,aACJ,mBAAmB,KAAK,OAAO,GAC3B,aAAa,KAAK,OAAO,GACzB,KAAK;AAEX,UAAO,GAAG,KAAK,KAAK,IAAI,KAAK,YAAY,UAAU,KAAK,UACtD,YAAY,WACb;IACD,CACD,KAAK,KAAK;;;;;;;;;;;;CAaf,OAAO,aACL,OACA,MACA;EACA,MAAM,EACJ,SAAS,QACT,SAAS,QACT,iBAAiB,CAAC,SAAS,mBAAmB,EAC9C,uBAAuB,iCACvB,gBAAgB,EAAE,KAChB,QAAQ,EAAE;EAEd,MAAM,WAAW;GACf,IAAI,4BACF,IAAI,eAAe;IACjB,UAJW;KAAC;KAAQ;KAAqB;KAAO,CAAC,KAAK,OAAO;IAK7D;IACA,kBAAkB;KAChB,cAAc,oBAAoB,wBAAwB,MAAM;KAChE,YAAY,MAAM,KAAK,SAAS,KAAK,KAAK,CAAC,KAAK,KAAK;KACtD;IACF,CAAC,CACH;GACD,GAAG;GACH,IAAI,2BACF,IAAI,eAAe;IACjB,UAAU;IACV;IACD,CAAC,CACH;GACF;AACD,SAAO,mBAAmB,aAAa,SAAS;;;;;;;;;;;CAYlD,OAAO,gBACL,KACA,OACA,MACA;AACA,sBAAoB,cAAc,MAAM;EACxC,MAAM,SAAS,oBAAoB,aAAa,OAAO,KAAK;EAC5D,MAAM,eACJ,MAAM,gBACN,oBAAoB,uBAAuB;GACzC;GACA,WAAW,MAAM,KAAK,SAAS,KAAK,KAAK;GAC1C,CAAC;AAOJ,SAAO,IAAI,oBAAoB;GAC7B,UAPY,IAAI,SAAS;IACzB;IACA;IACA,WAAW,MAAM;IAClB,CAAC;GAIA;GACA,cAAc,MAAM,KAAK,MAAM,EAAE,KAAK;GACvC,CAAC;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAgFN,eAAsB,0BAA0B,EAC9C,KACA,OACA,QACA,kBACkC;CAClC,MAAM,mBAAmB;EAAC;EAAS;EAAc;EAAmB,CAAC,QAClE,MAAM,CAAC,OAAO,eAAe,SAAS,EAAE,CAC1C;AACD,KAAI,iBAAiB,SAAS,EAC5B,OAAM,IAAI,MACR,wDAAwD,KAAK,UAC3D,iBACD,GACF;CAEH,IAAI,YAAsB,EAAE;AAC5B,KAAI,MAAM,MAAM,aAAa,CAC3B,aAAY,MAAM,KAAK,SAAS,KAAK,SAAS,KAAK;UAC1C,MAAM,MAAM,iBAAiB,CACtC,aAAY,MAAM,KAAK,SAAS,KAAK,KAAK;KAE1C,OAAM,IAAI,MACR,kEACD;CAEH,MAAM,kBAAkB,MAAM,OAAO,QAAQ;EAC3C,OAAO,6BAA6B,MAAM;EAC1C,YAAY,UAAU,KAAK,KAAK;EACjC,CAAC;CAEF,MAAM,cAAe,IAA0B,WAAW,EACxD,MAAM,CAAC,cAAc,EACtB,CAAC;AAmBF,QAlBc,sBAAsB,cAClC;EACE,oBAAoB,OAAO,EACzB,mBAAmB,UACjB,kBAAkB,MAAM,MAAM,EACjC,CAAC;EACF;EACA;EACA,sCAAsC,QAAQ,KAAK,EACjD,WACD,CAAC;EACH,EACD;EACE,MAAM;EACN;EACA,cAAc;EACf,CACF"}

package/dist/agents/structured_chat/outputParser.cjs

@@ -1,9 +1,9 @@
-const require_rolldown_runtime = require('../../_virtual/rolldown_runtime.cjs');
+const require_runtime = require('../../_virtual/_rolldown/runtime.cjs');
 const require_types = require('../types.cjs');
 const require_fix = require('../../output_parsers/fix.cjs');
 const require_prompt = require('./prompt.cjs');
-const __langchain_core_prompts = require_rolldown_runtime.__toESM(require("@langchain/core/prompts"));
-const __langchain_core_output_parsers = require_rolldown_runtime.__toESM(require("@langchain/core/output_parsers"));
+let _langchain_core_prompts = require("@langchain/core/prompts");
+let _langchain_core_output_parsers = require("@langchain/core/output_parsers");
 
 //#region src/agents/structured_chat/outputParser.ts
 /**
@@ -33,11 +33,9 @@ var StructuredChatOutputParser = class extends require_types.AgentActionOutputPa
 	*/
 	async parse(text) {
 		try {
-			const regex = /```(?:json)?(.*)(```)/gs;
-			const actionMatch = regex.exec(text);
-			if (actionMatch === null) throw new __langchain_core_output_parsers.OutputParserException(`Could not parse an action. The agent action must be within a markdown code block, and "action" must be a provided tool or "Final Answer"`);
-			const response = JSON.parse(actionMatch[1].trim());
-			const { action, action_input } = response;
+			const actionMatch = /```(?:json)?(.*)(```)/gs.exec(text);
+			if (actionMatch === null) throw new _langchain_core_output_parsers.OutputParserException(`Could not parse an action. The agent action must be within a markdown code block, and "action" must be a provided tool or "Final Answer"`);
+			const { action, action_input } = JSON.parse(actionMatch[1].trim());
 			if (action === "Final Answer") return {
 				returnValues: { output: action_input },
 				log: text
@@ -48,7 +46,7 @@ var StructuredChatOutputParser = class extends require_types.AgentActionOutputPa
 				log: text
 			};
 		} catch (e) {
-			throw new __langchain_core_output_parsers.OutputParserException(`Failed to parse. Text: "${text}". Error: ${e}`);
+			throw new _langchain_core_output_parsers.OutputParserException(`Failed to parse. Text: "${text}". Error: ${e}`);
 		}
 	}
 	/**
@@ -57,7 +55,7 @@ var StructuredChatOutputParser = class extends require_types.AgentActionOutputPa
 	* @returns A string representing the format instructions.
 	*/
 	getFormatInstructions() {
-		return (0, __langchain_core_prompts.renderTemplate)(require_prompt.AGENT_ACTION_FORMAT_INSTRUCTIONS, "f-string", { tool_names: this.toolNames.join(", ") });
+		return (0, _langchain_core_prompts.renderTemplate)(require_prompt.AGENT_ACTION_FORMAT_INSTRUCTIONS, "f-string", { tool_names: this.toolNames.join(", ") });
 	}
 };
 /**
@@ -109,7 +107,7 @@ var StructuredChatOutputParserWithRetries = class StructuredChatOutputParserWith
 	* @returns A string representing the format instructions.
 	*/
 	getFormatInstructions() {
-		return (0, __langchain_core_prompts.renderTemplate)(require_prompt.FORMAT_INSTRUCTIONS, "f-string", { tool_names: this.toolNames.join(", ") });
+		return (0, _langchain_core_prompts.renderTemplate)(require_prompt.FORMAT_INSTRUCTIONS, "f-string", { tool_names: this.toolNames.join(", ") });
 	}
 	/**
 	* Creates a new `StructuredChatOutputParserWithRetries` instance from a
@@ -121,10 +119,9 @@ var StructuredChatOutputParserWithRetries = class StructuredChatOutputParserWith
 	*/
 	static fromLLM(llm, options) {
 		const baseParser = options.baseParser ?? new StructuredChatOutputParser({ toolNames: options.toolNames ?? [] });
-		const outputFixingParser = require_fix.OutputFixingParser.fromLLM(llm, baseParser);
 		return new StructuredChatOutputParserWithRetries({
 			baseParser,
-			outputFixingParser,
+			outputFixingParser: require_fix.OutputFixingParser.fromLLM(llm, baseParser),
 			toolNames: options.toolNames
 		});
 	}

package/dist/agents/structured_chat/outputParser.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"outputParser.cjs","names":["AgentActionOutputParser","fields: { toolNames: string[] }","text: string","OutputParserException","AGENT_ACTION_FORMAT_INSTRUCTIONS","fields: StructuredChatOutputParserArgs","callbacks?: Callbacks","FORMAT_INSTRUCTIONS","llm: BaseLanguageModelInterface","options: Omit<StructuredChatOutputParserArgs, \"outputFixingParser\">","OutputFixingParser"],"sources":["../../../src/agents/structured_chat/outputParser.ts"],"sourcesContent":["import type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport { Callbacks } from \"@langchain/core/callbacks/manager\";\nimport { AgentAction, AgentFinish } from \"@langchain/core/agents\";\nimport { OutputParserException } from \"@langchain/core/output_parsers\";\nimport { renderTemplate } from \"@langchain/core/prompts\";\nimport { AgentActionOutputParser } from \"../types.js\";\nimport {\n  AGENT_ACTION_FORMAT_INSTRUCTIONS,\n  FORMAT_INSTRUCTIONS,\n} from \"./prompt.js\";\nimport { OutputFixingParser } from \"../../output_parsers/fix.js\";\n\n/**\n * A class that provides a custom implementation for parsing the output of\n * a StructuredChatAgent action. It extends the `AgentActionOutputParser`\n * class and extracts the action and action input from the text output,\n * returning an `AgentAction` or `AgentFinish` object.\n */\nexport class StructuredChatOutputParser extends AgentActionOutputParser {\n  lc_namespace = [\"langchain\", \"agents\", \"structured_chat\"];\n\n  private toolNames: string[];\n\n  constructor(fields: { toolNames: string[] }) {\n    super(...arguments);\n    this.toolNames = fields.toolNames;\n  }\n\n  /**\n   * Parses the given text and returns an `AgentAction` or `AgentFinish`\n   * object. If an `OutputFixingParser` is provided, it is used for parsing;\n   * otherwise, the base parser is used.\n   * @param text The text to parse.\n   * @param callbacks Optional callbacks for asynchronous operations.\n   * @returns A Promise that resolves to an `AgentAction` or `AgentFinish` object.\n   */\n  async parse(text: string): Promise<AgentAction | AgentFinish> {\n    try {\n      const regex = /```(?:json)?(.*)(```)/gs;\n      const actionMatch = regex.exec(text);\n      if (actionMatch === null) {\n        throw new OutputParserException(\n          `Could not parse an action. The agent action must be within a markdown code block, and \"action\" must be a provided tool or \"Final Answer\"`\n        );\n      }\n      const response = JSON.parse(actionMatch[1].trim());\n      const { action, action_input } = response;\n\n      if (action === \"Final Answer\") {\n        return { returnValues: { output: action_input }, log: text };\n      }\n      return { tool: action, toolInput: action_input || {}, log: text };\n    } catch (e) {\n      throw new OutputParserException(\n        `Failed to parse. Text: \"${text}\". Error: ${e}`\n      );\n    }\n  }\n\n  /**\n   * Returns the format instructions for parsing the output of an agent\n   * action in the style of the StructuredChatAgent.\n   * @returns A string representing the format instructions.\n   */\n  getFormatInstructions(): string {\n    return renderTemplate(AGENT_ACTION_FORMAT_INSTRUCTIONS, \"f-string\", {\n      tool_names: this.toolNames.join(\", \"),\n    });\n  }\n}\n\n/**\n * An interface for the arguments used to construct a\n * `StructuredChatOutputParserWithRetries` instance.\n */\nexport interface StructuredChatOutputParserArgs {\n  baseParser?: StructuredChatOutputParser;\n  outputFixingParser?: OutputFixingParser<AgentAction | AgentFinish>;\n  toolNames?: string[];\n}\n\n/**\n * A class that provides a wrapper around the `StructuredChatOutputParser`\n * and `OutputFixingParser` classes. It extends the\n * `AgentActionOutputParser` class and allows for retrying the output\n * parsing using the `OutputFixingParser` if it is provided.\n * @example\n * ```typescript\n * const outputParser = new StructuredChatOutputParserWithRetries.fromLLM(\n *   new ChatOpenAI({ model: \"gpt-4o-mini\", temperature: 0 }),\n *   {\n *     toolNames: [\"calculator\", \"random-number-generator\"],\n *   },\n * );\n * const result = await outputParser.parse(\n *  \"What is a random number between 5 and 10 raised to the second power?\"\n * );\n * ```\n */\nexport class StructuredChatOutputParserWithRetries extends AgentActionOutputParser {\n  lc_namespace = [\"langchain\", \"agents\", \"structured_chat\"];\n\n  private baseParser: StructuredChatOutputParser;\n\n  private outputFixingParser?: OutputFixingParser<AgentAction | AgentFinish>;\n\n  private toolNames: string[] = [];\n\n  constructor(fields: StructuredChatOutputParserArgs) {\n    super(fields);\n    this.toolNames = fields.toolNames ?? this.toolNames;\n    this.baseParser =\n      fields?.baseParser ??\n      new StructuredChatOutputParser({ toolNames: this.toolNames });\n    this.outputFixingParser = fields?.outputFixingParser;\n  }\n\n  /**\n   * Parses the given text and returns an `AgentAction` or `AgentFinish`\n   * object. Throws an `OutputParserException` if the parsing fails.\n   * @param text The text to parse.\n   * @returns A Promise that resolves to an `AgentAction` or `AgentFinish` object.\n   */\n  async parse(\n    text: string,\n    callbacks?: Callbacks\n  ): Promise<AgentAction | AgentFinish> {\n    if (this.outputFixingParser !== undefined) {\n      return this.outputFixingParser.parse(text, callbacks);\n    }\n    return this.baseParser.parse(text);\n  }\n\n  /**\n   * Returns the format instructions for parsing the output of an agent\n   * action in the style of the StructuredChatAgent.\n   * @returns A string representing the format instructions.\n   */\n  getFormatInstructions(): string {\n    return renderTemplate(FORMAT_INSTRUCTIONS, \"f-string\", {\n      tool_names: this.toolNames.join(\", \"),\n    });\n  }\n\n  /**\n   * Creates a new `StructuredChatOutputParserWithRetries` instance from a\n   * `BaseLanguageModel` and options. The options can include a base parser\n   * and tool names.\n   * @param llm A `BaseLanguageModel` instance.\n   * @param options Options for creating a `StructuredChatOutputParserWithRetries` instance.\n   * @returns A new `StructuredChatOutputParserWithRetries` instance.\n   */\n  static fromLLM(\n    llm: BaseLanguageModelInterface,\n    options: Omit<StructuredChatOutputParserArgs, \"outputFixingParser\">\n  ): StructuredChatOutputParserWithRetries {\n    const baseParser =\n      options.baseParser ??\n      new StructuredChatOutputParser({ toolNames: options.toolNames ?? [] });\n    const outputFixingParser = OutputFixingParser.fromLLM(llm, baseParser);\n    return new StructuredChatOutputParserWithRetries({\n      baseParser,\n      outputFixingParser,\n      toolNames: options.toolNames,\n    });\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;AAkBA,IAAa,6BAAb,cAAgDA,sCAAwB;CACtE,eAAe;EAAC;EAAa;EAAU;CAAkB;CAEzD,AAAQ;CAER,YAAYC,QAAiC;EAC3C,MAAM,GAAG,UAAU;EACnB,KAAK,YAAY,OAAO;CACzB;;;;;;;;;CAUD,MAAM,MAAMC,MAAkD;AAC5D,MAAI;GACF,MAAM,QAAQ;GACd,MAAM,cAAc,MAAM,KAAK,KAAK;AACpC,OAAI,gBAAgB,KAClB,OAAM,IAAIC,sDACR,CAAC,wIAAwI,CAAC;GAG9I,MAAM,WAAW,KAAK,MAAM,YAAY,GAAG,MAAM,CAAC;GAClD,MAAM,EAAE,QAAQ,cAAc,GAAG;AAEjC,OAAI,WAAW,eACb,QAAO;IAAE,cAAc,EAAE,QAAQ,aAAc;IAAE,KAAK;GAAM;AAE9D,UAAO;IAAE,MAAM;IAAQ,WAAW,gBAAgB,CAAE;IAAE,KAAK;GAAM;EAClE,SAAQ,GAAG;AACV,SAAM,IAAIA,sDACR,CAAC,wBAAwB,EAAE,KAAK,UAAU,EAAE,GAAG;EAElD;CACF;;;;;;CAOD,wBAAgC;AAC9B,sDAAsBC,iDAAkC,YAAY,EAClE,YAAY,KAAK,UAAU,KAAK,KAAK,CACtC,EAAC;CACH;AACF;;;;;;;;;;;;;;;;;;;AA8BD,IAAa,wCAAb,MAAa,8CAA8CJ,sCAAwB;CACjF,eAAe;EAAC;EAAa;EAAU;CAAkB;CAEzD,AAAQ;CAER,AAAQ;CAER,AAAQ,YAAsB,CAAE;CAEhC,YAAYK,QAAwC;EAClD,MAAM,OAAO;EACb,KAAK,YAAY,OAAO,aAAa,KAAK;EAC1C,KAAK,aACH,QAAQ,cACR,IAAI,2BAA2B,EAAE,WAAW,KAAK,UAAW;EAC9D,KAAK,qBAAqB,QAAQ;CACnC;;;;;;;CAQD,MAAM,MACJH,MACAI,WACoC;AACpC,MAAI,KAAK,uBAAuB,OAC9B,QAAO,KAAK,mBAAmB,MAAM,MAAM,UAAU;AAEvD,SAAO,KAAK,WAAW,MAAM,KAAK;CACnC;;;;;;CAOD,wBAAgC;AAC9B,sDAAsBC,oCAAqB,YAAY,EACrD,YAAY,KAAK,UAAU,KAAK,KAAK,CACtC,EAAC;CACH;;;;;;;;;CAUD,OAAO,QACLC,KACAC,SACuC;EACvC,MAAM,aACJ,QAAQ,cACR,IAAI,2BAA2B,EAAE,WAAW,QAAQ,aAAa,CAAE,EAAE;EACvE,MAAM,qBAAqBC,+BAAmB,QAAQ,KAAK,WAAW;AACtE,SAAO,IAAI,sCAAsC;GAC/C;GACA;GACA,WAAW,QAAQ;EACpB;CACF;AACF"}
+{"version":3,"file":"outputParser.cjs","names":["AgentActionOutputParser","OutputParserException","AGENT_ACTION_FORMAT_INSTRUCTIONS","FORMAT_INSTRUCTIONS","OutputFixingParser"],"sources":["../../../src/agents/structured_chat/outputParser.ts"],"sourcesContent":["import type { BaseLanguageModelInterface } from \"@langchain/core/language_models/base\";\nimport { Callbacks } from \"@langchain/core/callbacks/manager\";\nimport { AgentAction, AgentFinish } from \"@langchain/core/agents\";\nimport { OutputParserException } from \"@langchain/core/output_parsers\";\nimport { renderTemplate } from \"@langchain/core/prompts\";\nimport { AgentActionOutputParser } from \"../types.js\";\nimport {\n  AGENT_ACTION_FORMAT_INSTRUCTIONS,\n  FORMAT_INSTRUCTIONS,\n} from \"./prompt.js\";\nimport { OutputFixingParser } from \"../../output_parsers/fix.js\";\n\n/**\n * A class that provides a custom implementation for parsing the output of\n * a StructuredChatAgent action. It extends the `AgentActionOutputParser`\n * class and extracts the action and action input from the text output,\n * returning an `AgentAction` or `AgentFinish` object.\n */\nexport class StructuredChatOutputParser extends AgentActionOutputParser {\n  lc_namespace = [\"langchain\", \"agents\", \"structured_chat\"];\n\n  private toolNames: string[];\n\n  constructor(fields: { toolNames: string[] }) {\n    super(...arguments);\n    this.toolNames = fields.toolNames;\n  }\n\n  /**\n   * Parses the given text and returns an `AgentAction` or `AgentFinish`\n   * object. If an `OutputFixingParser` is provided, it is used for parsing;\n   * otherwise, the base parser is used.\n   * @param text The text to parse.\n   * @param callbacks Optional callbacks for asynchronous operations.\n   * @returns A Promise that resolves to an `AgentAction` or `AgentFinish` object.\n   */\n  async parse(text: string): Promise<AgentAction | AgentFinish> {\n    try {\n      const regex = /```(?:json)?(.*)(```)/gs;\n      const actionMatch = regex.exec(text);\n      if (actionMatch === null) {\n        throw new OutputParserException(\n          `Could not parse an action. The agent action must be within a markdown code block, and \"action\" must be a provided tool or \"Final Answer\"`\n        );\n      }\n      const response = JSON.parse(actionMatch[1].trim());\n      const { action, action_input } = response;\n\n      if (action === \"Final Answer\") {\n        return { returnValues: { output: action_input }, log: text };\n      }\n      return { tool: action, toolInput: action_input || {}, log: text };\n    } catch (e) {\n      throw new OutputParserException(\n        `Failed to parse. Text: \"${text}\". Error: ${e}`\n      );\n    }\n  }\n\n  /**\n   * Returns the format instructions for parsing the output of an agent\n   * action in the style of the StructuredChatAgent.\n   * @returns A string representing the format instructions.\n   */\n  getFormatInstructions(): string {\n    return renderTemplate(AGENT_ACTION_FORMAT_INSTRUCTIONS, \"f-string\", {\n      tool_names: this.toolNames.join(\", \"),\n    });\n  }\n}\n\n/**\n * An interface for the arguments used to construct a\n * `StructuredChatOutputParserWithRetries` instance.\n */\nexport interface StructuredChatOutputParserArgs {\n  baseParser?: StructuredChatOutputParser;\n  outputFixingParser?: OutputFixingParser<AgentAction | AgentFinish>;\n  toolNames?: string[];\n}\n\n/**\n * A class that provides a wrapper around the `StructuredChatOutputParser`\n * and `OutputFixingParser` classes. It extends the\n * `AgentActionOutputParser` class and allows for retrying the output\n * parsing using the `OutputFixingParser` if it is provided.\n * @example\n * ```typescript\n * const outputParser = new StructuredChatOutputParserWithRetries.fromLLM(\n *   new ChatOpenAI({ model: \"gpt-4o-mini\", temperature: 0 }),\n *   {\n *     toolNames: [\"calculator\", \"random-number-generator\"],\n *   },\n * );\n * const result = await outputParser.parse(\n *  \"What is a random number between 5 and 10 raised to the second power?\"\n * );\n * ```\n */\nexport class StructuredChatOutputParserWithRetries extends AgentActionOutputParser {\n  lc_namespace = [\"langchain\", \"agents\", \"structured_chat\"];\n\n  private baseParser: StructuredChatOutputParser;\n\n  private outputFixingParser?: OutputFixingParser<AgentAction | AgentFinish>;\n\n  private toolNames: string[] = [];\n\n  constructor(fields: StructuredChatOutputParserArgs) {\n    super(fields);\n    this.toolNames = fields.toolNames ?? this.toolNames;\n    this.baseParser =\n      fields?.baseParser ??\n      new StructuredChatOutputParser({ toolNames: this.toolNames });\n    this.outputFixingParser = fields?.outputFixingParser;\n  }\n\n  /**\n   * Parses the given text and returns an `AgentAction` or `AgentFinish`\n   * object. Throws an `OutputParserException` if the parsing fails.\n   * @param text The text to parse.\n   * @returns A Promise that resolves to an `AgentAction` or `AgentFinish` object.\n   */\n  async parse(\n    text: string,\n    callbacks?: Callbacks\n  ): Promise<AgentAction | AgentFinish> {\n    if (this.outputFixingParser !== undefined) {\n      return this.outputFixingParser.parse(text, callbacks);\n    }\n    return this.baseParser.parse(text);\n  }\n\n  /**\n   * Returns the format instructions for parsing the output of an agent\n   * action in the style of the StructuredChatAgent.\n   * @returns A string representing the format instructions.\n   */\n  getFormatInstructions(): string {\n    return renderTemplate(FORMAT_INSTRUCTIONS, \"f-string\", {\n      tool_names: this.toolNames.join(\", \"),\n    });\n  }\n\n  /**\n   * Creates a new `StructuredChatOutputParserWithRetries` instance from a\n   * `BaseLanguageModel` and options. The options can include a base parser\n   * and tool names.\n   * @param llm A `BaseLanguageModel` instance.\n   * @param options Options for creating a `StructuredChatOutputParserWithRetries` instance.\n   * @returns A new `StructuredChatOutputParserWithRetries` instance.\n   */\n  static fromLLM(\n    llm: BaseLanguageModelInterface,\n    options: Omit<StructuredChatOutputParserArgs, \"outputFixingParser\">\n  ): StructuredChatOutputParserWithRetries {\n    const baseParser =\n      options.baseParser ??\n      new StructuredChatOutputParser({ toolNames: options.toolNames ?? [] });\n    const outputFixingParser = OutputFixingParser.fromLLM(llm, baseParser);\n    return new StructuredChatOutputParserWithRetries({\n      baseParser,\n      outputFixingParser,\n      toolNames: options.toolNames,\n    });\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;AAkBA,IAAa,6BAAb,cAAgDA,sCAAwB;CACtE,eAAe;EAAC;EAAa;EAAU;EAAkB;CAEzD,AAAQ;CAER,YAAY,QAAiC;AAC3C,QAAM,GAAG,UAAU;AACnB,OAAK,YAAY,OAAO;;;;;;;;;;CAW1B,MAAM,MAAM,MAAkD;AAC5D,MAAI;GAEF,MAAM,cADQ,0BACY,KAAK,KAAK;AACpC,OAAI,gBAAgB,KAClB,OAAM,IAAIC,qDACR,2IACD;GAGH,MAAM,EAAE,QAAQ,iBADC,KAAK,MAAM,YAAY,GAAG,MAAM,CAAC;AAGlD,OAAI,WAAW,eACb,QAAO;IAAE,cAAc,EAAE,QAAQ,cAAc;IAAE,KAAK;IAAM;AAE9D,UAAO;IAAE,MAAM;IAAQ,WAAW,gBAAgB,EAAE;IAAE,KAAK;IAAM;WAC1D,GAAG;AACV,SAAM,IAAIA,qDACR,2BAA2B,KAAK,YAAY,IAC7C;;;;;;;;CASL,wBAAgC;AAC9B,qDAAsBC,iDAAkC,YAAY,EAClE,YAAY,KAAK,UAAU,KAAK,KAAK,EACtC,CAAC;;;;;;;;;;;;;;;;;;;;;AAgCN,IAAa,wCAAb,MAAa,8CAA8CF,sCAAwB;CACjF,eAAe;EAAC;EAAa;EAAU;EAAkB;CAEzD,AAAQ;CAER,AAAQ;CAER,AAAQ,YAAsB,EAAE;CAEhC,YAAY,QAAwC;AAClD,QAAM,OAAO;AACb,OAAK,YAAY,OAAO,aAAa,KAAK;AAC1C,OAAK,aACH,QAAQ,cACR,IAAI,2BAA2B,EAAE,WAAW,KAAK,WAAW,CAAC;AAC/D,OAAK,qBAAqB,QAAQ;;;;;;;;CASpC,MAAM,MACJ,MACA,WACoC;AACpC,MAAI,KAAK,uBAAuB,OAC9B,QAAO,KAAK,mBAAmB,MAAM,MAAM,UAAU;AAEvD,SAAO,KAAK,WAAW,MAAM,KAAK;;;;;;;CAQpC,wBAAgC;AAC9B,qDAAsBG,oCAAqB,YAAY,EACrD,YAAY,KAAK,UAAU,KAAK,KAAK,EACtC,CAAC;;;;;;;;;;CAWJ,OAAO,QACL,KACA,SACuC;EACvC,MAAM,aACJ,QAAQ,cACR,IAAI,2BAA2B,EAAE,WAAW,QAAQ,aAAa,EAAE,EAAE,CAAC;AAExE,SAAO,IAAI,sCAAsC;GAC/C;GACA,oBAHyBC,+BAAmB,QAAQ,KAAK,WAAW;GAIpE,WAAW,QAAQ;GACpB,CAAC"}

package/dist/agents/structured_chat/outputParser.d.cts

@@ -5,7 +5,6 @@ import { BaseLanguageModelInterface } from "@langchain/core/language_models/base
 import { Callbacks } from "@langchain/core/callbacks/manager";
 
 //#region src/agents/structured_chat/outputParser.d.ts
-
 /**
  * A class that provides a custom implementation for parsing the output of
  * a StructuredChatAgent action. It extends the `AgentActionOutputParser`