n8n-io · netroy · Dec 16, 2024
diff --git a/packages/@n8n/nodes-langchain/.eslintrc.js b/packages/@n8n/nodes-langchain/.eslintrc.js
@@ -15,7 +15,6 @@ module.exports = {
 		eqeqeq: 'warn',
 		'id-denylist': 'warn',
 		'import/extensions': 'warn',
-		'import/order': 'warn',
 		'prefer-spread': 'warn',
 
 		'@typescript-eslint/naming-convention': ['error', { selector: 'memberLike', format: null }],

diff --git a/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/ConversationalAgent/execute.ts b/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/ConversationalAgent/execute.ts
@@ -1,10 +1,9 @@
-import type { BaseChatMemory } from '@langchain/community/memory/chat_memory';
 import type { BaseOutputParser } from '@langchain/core/output_parsers';
 import { PromptTemplate } from '@langchain/core/prompts';
 import { initializeAgentExecutorWithOptions } from 'langchain/agents';
 import { CombiningOutputParser } from 'langchain/output_parsers';
 import type { IExecuteFunctions, INodeExecutionData } from 'n8n-workflow';
-import { NodeConnectionType, NodeOperationError } from 'n8n-workflow';
+import { NodeOperationError } from 'n8n-workflow';
 
 import { isChatInstance, getPromptInputByType, getConnectedTools } from '@utils/helpers';
 import { getOptionalOutputParsers } from '@utils/output_parsers/N8nOutputParser';
@@ -18,15 +17,13 @@ export async function conversationalAgentExecute(
 	nodeVersion: number,
 ): Promise<INodeExecutionData[][]> {
 	this.logger.debug('Executing Conversational Agent');
-	const model = await this.getInputConnectionData(NodeConnectionType.AiLanguageModel, 0);
+	const model = await this.aiRootNodeContext.getModel();
 
 	if (!isChatInstance(model)) {
 		throw new NodeOperationError(this.getNode(), 'Conversational Agent requires Chat Model');
 	}
 
-	const memory = (await this.getInputConnectionData(NodeConnectionType.AiMemory, 0)) as
-		| BaseChatMemory
-		| undefined;
+	const memory = await this.aiRootNodeContext.getMemory();
 
 	const tools = await getConnectedTools(this, nodeVersion >= 1.5, true, true);
 	const outputParsers = await getOptionalOutputParsers(this);

diff --git a/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/OpenAiFunctionsAgent/execute.ts b/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/OpenAiFunctionsAgent/execute.ts
@@ -3,14 +3,9 @@ import { PromptTemplate } from '@langchain/core/prompts';
 import { ChatOpenAI } from '@langchain/openai';
 import type { AgentExecutorInput } from 'langchain/agents';
 import { AgentExecutor, OpenAIAgent } from 'langchain/agents';
-import { BufferMemory, type BaseChatMemory } from 'langchain/memory';
+import { BufferMemory } from 'langchain/memory';
 import { CombiningOutputParser } from 'langchain/output_parsers';
-import {
-	type IExecuteFunctions,
-	type INodeExecutionData,
-	NodeConnectionType,
-	NodeOperationError,
-} from 'n8n-workflow';
+import { type IExecuteFunctions, type INodeExecutionData, NodeOperationError } from 'n8n-workflow';
 
 import { getConnectedTools, getPromptInputByType } from '@utils/helpers';
 import { getOptionalOutputParsers } from '@utils/output_parsers/N8nOutputParser';
@@ -23,20 +18,15 @@ export async function openAiFunctionsAgentExecute(
 	nodeVersion: number,
 ): Promise<INodeExecutionData[][]> {
 	this.logger.debug('Executing OpenAi Functions Agent');
-	const model = (await this.getInputConnectionData(
-		NodeConnectionType.AiLanguageModel,
-		0,
-	)) as ChatOpenAI;
+	const model = await this.aiRootNodeContext.getModel();
 
 	if (!(model instanceof ChatOpenAI)) {
 		throw new NodeOperationError(
 			this.getNode(),
 			'OpenAI Functions Agent requires OpenAI Chat Model',
 		);
 	}
-	const memory = (await this.getInputConnectionData(NodeConnectionType.AiMemory, 0)) as
-		| BaseChatMemory
-		| undefined;
+	const memory = await this.aiRootNodeContext.getMemory();
 	const tools = await getConnectedTools(this, nodeVersion >= 1.5, false);
 	const outputParsers = await getOptionalOutputParsers(this);
 	const options = this.getNodeParameter('options', 0, {}) as {

diff --git a/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/PlanAndExecuteAgent/execute.ts b/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/PlanAndExecuteAgent/execute.ts
@@ -1,14 +1,8 @@
-import type { BaseChatModel } from '@langchain/core/language_models/chat_models';
 import type { BaseOutputParser } from '@langchain/core/output_parsers';
 import { PromptTemplate } from '@langchain/core/prompts';
 import { PlanAndExecuteAgentExecutor } from 'langchain/experimental/plan_and_execute';
 import { CombiningOutputParser } from 'langchain/output_parsers';
-import {
-	type IExecuteFunctions,
-	type INodeExecutionData,
-	NodeConnectionType,
-	NodeOperationError,
-} from 'n8n-workflow';
+import { type IExecuteFunctions, type INodeExecutionData, NodeOperationError } from 'n8n-workflow';
 
 import { getConnectedTools, getPromptInputByType } from '@utils/helpers';
 import { getOptionalOutputParsers } from '@utils/output_parsers/N8nOutputParser';
@@ -22,11 +16,7 @@ export async function planAndExecuteAgentExecute(
 	nodeVersion: number,
 ): Promise<INodeExecutionData[][]> {
 	this.logger.debug('Executing PlanAndExecute Agent');
-	const model = (await this.getInputConnectionData(
-		NodeConnectionType.AiLanguageModel,
-		0,
-	)) as BaseChatModel;
-
+	const model = await this.aiRootNodeContext.getModel();
 	const tools = await getConnectedTools(this, nodeVersion >= 1.5, true, true);
 
 	await checkForStructuredTools(tools, this.getNode(), 'Plan & Execute Agent');

diff --git a/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/ReActAgent/execute.ts b/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/ReActAgent/execute.ts
@@ -1,15 +1,8 @@
-import type { BaseLanguageModel } from '@langchain/core/language_models/base';
-import type { BaseChatModel } from '@langchain/core/language_models/chat_models';
 import type { BaseOutputParser } from '@langchain/core/output_parsers';
 import { PromptTemplate } from '@langchain/core/prompts';
 import { AgentExecutor, ChatAgent, ZeroShotAgent } from 'langchain/agents';
 import { CombiningOutputParser } from 'langchain/output_parsers';
-import {
-	type IExecuteFunctions,
-	type INodeExecutionData,
-	NodeConnectionType,
-	NodeOperationError,
-} from 'n8n-workflow';
+import { type IExecuteFunctions, type INodeExecutionData, NodeOperationError } from 'n8n-workflow';
 
 import { getConnectedTools, getPromptInputByType, isChatInstance } from '@utils/helpers';
 import { getOptionalOutputParsers } from '@utils/output_parsers/N8nOutputParser';
@@ -24,9 +17,7 @@ export async function reActAgentAgentExecute(
 ): Promise<INodeExecutionData[][]> {
 	this.logger.debug('Executing ReAct Agent');
 
-	const model = (await this.getInputConnectionData(NodeConnectionType.AiLanguageModel, 0)) as
-		| BaseLanguageModel
-		| BaseChatModel;
+	const model = await this.aiRootNodeContext.getModel();
 
 	const tools = await getConnectedTools(this, nodeVersion >= 1.5, true, true);
 

diff --git a/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/SqlAgent/execute.ts b/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/SqlAgent/execute.ts
@@ -1,13 +1,10 @@
-import type { BaseChatMemory } from '@langchain/community/memory/chat_memory';
-import type { BaseLanguageModel } from '@langchain/core/language_models/base';
 import type { DataSource } from '@n8n/typeorm';
 import type { SqlCreatePromptArgs } from 'langchain/agents/toolkits/sql';
 import { SqlToolkit, createSqlAgent } from 'langchain/agents/toolkits/sql';
 import { SqlDatabase } from 'langchain/sql_db';
 import {
 	type IExecuteFunctions,
 	type INodeExecutionData,
-	NodeConnectionType,
 	NodeOperationError,
 	type IDataObject,
 } from 'n8n-workflow';
@@ -31,10 +28,8 @@ export async function sqlAgentAgentExecute(
 ): Promise<INodeExecutionData[][]> {
 	this.logger.debug('Executing SQL Agent');
 
-	const model = (await this.getInputConnectionData(
-		NodeConnectionType.AiLanguageModel,
-		0,
-	)) as BaseLanguageModel;
+	const model = await this.aiRootNodeContext.getModel();
+
 	const items = this.getInputData();
 
 	const returnData: INodeExecutionData[] = [];
@@ -113,9 +108,7 @@ export async function sqlAgentAgentExecute(
 			const toolkit = new SqlToolkit(dbInstance, model);
 			const agentExecutor = createSqlAgent(model, toolkit, agentOptions);
 
-			const memory = (await this.getInputConnectionData(NodeConnectionType.AiMemory, 0)) as
-				| BaseChatMemory
-				| undefined;
+			const memory = await this.aiRootNodeContext.getMemory();
 
 			agentExecutor.memory = memory;
 

diff --git a/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/ToolsAgent/execute.ts b/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/ToolsAgent/execute.ts
@@ -1,4 +1,3 @@
-import type { BaseChatMemory } from '@langchain/community/memory/chat_memory';
 import { HumanMessage } from '@langchain/core/messages';
 import type { BaseMessage } from '@langchain/core/messages';
 import type { BaseMessagePromptTemplateLike } from '@langchain/core/prompts';
@@ -9,9 +8,8 @@ import { DynamicStructuredTool } from '@langchain/core/tools';
 import type { AgentAction, AgentFinish } from 'langchain/agents';
 import { AgentExecutor, createToolCallingAgent } from 'langchain/agents';
 import { omit } from 'lodash';
-import { BINARY_ENCODING, jsonParse, NodeConnectionType, NodeOperationError } from 'n8n-workflow';
-import type { IExecuteFunctions, INodeExecutionData } from 'n8n-workflow';
-import type { ZodObject } from 'zod';
+import { BINARY_ENCODING, jsonParse, NodeOperationError } from 'n8n-workflow';
+import type { IExecuteFunctions, INodeExecutionData, ZodObjectAny } from 'n8n-workflow';
 import { z } from 'zod';
 
 import { isChatInstance, getPromptInputByType, getConnectedTools } from '@utils/helpers';
@@ -22,9 +20,8 @@ import {
 
 import { SYSTEM_MESSAGE } from './prompt';
 
-function getOutputParserSchema(outputParser: N8nOutputParser): ZodObject<any, any, any, any> {
-	const schema =
-		(outputParser.getSchema() as ZodObject<any, any, any, any>) ?? z.object({ text: z.string() });
+function getOutputParserSchema(outputParser: N8nOutputParser): ZodObjectAny {
+	const schema = (outputParser.getSchema() as ZodObjectAny) ?? z.object({ text: z.string() });
 
 	return schema;
 }
@@ -98,7 +95,7 @@ function fixEmptyContentMessage(steps: AgentFinish | AgentAction[]) {
 
 export async function toolsAgentExecute(this: IExecuteFunctions): Promise<INodeExecutionData[][]> {
 	this.logger.debug('Executing Tools Agent');
-	const model = await this.getInputConnectionData(NodeConnectionType.AiLanguageModel, 0);
+	const model = await this.aiRootNodeContext.getModel();
 
 	if (!isChatInstance(model) || !model.bindTools) {
 		throw new NodeOperationError(
@@ -107,9 +104,7 @@ export async function toolsAgentExecute(this: IExecuteFunctions): Promise<INodeE
 		);
 	}
 
-	const memory = (await this.getInputConnectionData(NodeConnectionType.AiMemory, 0)) as
-		| BaseChatMemory
-		| undefined;
+	const memory = await this.aiRootNodeContext.getMemory();
 
 	const tools = (await getConnectedTools(this, true, false)) as Array<DynamicStructuredTool | Tool>;
 	const outputParser = (await getOptionalOutputParsers(this))?.[0];

diff --git a/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/utils.ts b/packages/@n8n/nodes-langchain/nodes/agents/Agent/agents/utils.ts
@@ -1,9 +1,7 @@
 import type { BaseOutputParser } from '@langchain/core/output_parsers';
 import type { DynamicStructuredTool, Tool } from 'langchain/tools';
-import { NodeOperationError, type IExecuteFunctions, type INode } from 'n8n-workflow';
-import type { z } from 'zod';
-
-type ZodObjectAny = z.ZodObject<any, any, any, any>;
+import type { IExecuteFunctions, INode, ZodObjectAny } from 'n8n-workflow';
+import { NodeOperationError } from 'n8n-workflow';
 
 export async function extractParsedOutput(
 	ctx: IExecuteFunctions,

diff --git a/packages/@n8n/nodes-langchain/nodes/chains/ChainLLM/ChainLlm.node.ts b/packages/@n8n/nodes-langchain/nodes/chains/ChainLLM/ChainLlm.node.ts
@@ -84,10 +84,8 @@ async function getImageMessage(
 	}
 
 	const bufferData = await context.helpers.getBinaryDataBuffer(itemIndex, binaryDataKey);
-	const model = (await context.getInputConnectionData(
-		NodeConnectionType.AiLanguageModel,
-		0,
-	)) as BaseLanguageModel;
+
+	const model = await context.aiRootNodeContext.getModel();
 	const dataURI = `data:image/jpeg;base64,${bufferData.toString('base64')}`;
 
 	const directUriModels = [ChatGoogleGenerativeAI, ChatOllama];
@@ -108,7 +106,7 @@ async function getImageMessage(
 async function getChainPromptTemplate(
 	context: IExecuteFunctions,
 	itemIndex: number,
-	llm: BaseLanguageModel | BaseChatModel,
+	model: BaseLanguageModel | BaseChatModel,
 	messages?: MessagesTemplate[],
 	formatInstructions?: string,
 	query?: string,
@@ -119,7 +117,7 @@ async function getChainPromptTemplate(
 		partialVariables: formatInstructions ? { formatInstructions } : undefined,
 	});
 
-	if (isChatInstance(llm)) {
+	if (isChatInstance(model)) {
 		const parsedMessages = await Promise.all(
 			(messages ?? []).map(async (message) => {
 				const messageClass = [
@@ -166,12 +164,12 @@ async function getChainPromptTemplate(
 
 async function createSimpleLLMChain(
 	context: IExecuteFunctions,
-	llm: BaseLanguageModel,
+	model: BaseLanguageModel,
 	query: string,
 	prompt: ChatPromptTemplate | PromptTemplate,
 ): Promise<string[]> {
 	const chain = new LLMChain({
-		llm,
+		llm: model,
 		prompt,
 	}).withConfig(getTracingConfig(context));
 
@@ -187,22 +185,22 @@ async function getChain(
 	context: IExecuteFunctions,
 	itemIndex: number,
 	query: string,
-	llm: BaseLanguageModel,
+	model: BaseLanguageModel,
 	outputParsers: N8nOutputParser[],
 	messages?: MessagesTemplate[],
 ): Promise<unknown[]> {
 	const chatTemplate: ChatPromptTemplate | PromptTemplate = await getChainPromptTemplate(
 		context,
 		itemIndex,
-		llm,
+		model,
 		messages,
 		undefined,
 		query,
 	);
 
 	// If there are no output parsers, create a simple LLM chain and execute the query
 	if (!outputParsers.length) {
-		return await createSimpleLLMChain(context, llm, query, chatTemplate);
+		return await createSimpleLLMChain(context, model, query, chatTemplate);
 	}
 
 	// If there's only one output parser, use it; otherwise, create a combined output parser
@@ -215,13 +213,13 @@ async function getChain(
 	const prompt = await getChainPromptTemplate(
 		context,
 		itemIndex,
-		llm,
+		model,
 		messages,
 		formatInstructions,
 		query,
 	);
 
-	const chain = prompt.pipe(llm).pipe(combinedOutputParser);
+	const chain = prompt.pipe(model).pipe(combinedOutputParser);
 	const response = (await chain.withConfig(getTracingConfig(context)).invoke({ query })) as
 		| string
 		| string[];
@@ -515,10 +513,8 @@ export class ChainLlm implements INodeType {
 		const items = this.getInputData();
 
 		const returnData: INodeExecutionData[] = [];
-		const llm = (await this.getInputConnectionData(
-			NodeConnectionType.AiLanguageModel,
-			0,
-		)) as BaseLanguageModel;
+
+		const model = await this.aiRootNodeContext.getModel();
 
 		const outputParsers = await getOptionalOutputParsers(this);
 
@@ -545,7 +541,7 @@ export class ChainLlm implements INodeType {
 					throw new NodeOperationError(this.getNode(), "The 'prompt' parameter is empty.");
 				}
 
-				const responses = await getChain(this, itemIndex, prompt, llm, outputParsers, messages);
+				const responses = await getChain(this, itemIndex, prompt, model, outputParsers, messages);
 
 				responses.forEach((response) => {
 					let data: IDataObject;

diff --git a/packages/@n8n/nodes-langchain/nodes/chains/ChainRetrievalQA/ChainRetrievalQa.node.ts b/packages/@n8n/nodes-langchain/nodes/chains/ChainRetrievalQA/ChainRetrievalQa.node.ts
@@ -1,11 +1,9 @@
-import type { BaseLanguageModel } from '@langchain/core/language_models/base';
 import {
 	ChatPromptTemplate,
 	SystemMessagePromptTemplate,
 	HumanMessagePromptTemplate,
 	PromptTemplate,
 } from '@langchain/core/prompts';
-import type { BaseRetriever } from '@langchain/core/retrievers';
 import { RetrievalQAChain } from 'langchain/chains';
 import {
 	NodeConnectionType,
@@ -161,15 +159,8 @@ export class ChainRetrievalQa implements INodeType {
 	async execute(this: IExecuteFunctions): Promise<INodeExecutionData[][]> {
 		this.logger.debug('Executing Retrieval QA Chain');
 
-		const model = (await this.getInputConnectionData(
-			NodeConnectionType.AiLanguageModel,
-			0,
-		)) as BaseLanguageModel;
-
-		const retriever = (await this.getInputConnectionData(
-			NodeConnectionType.AiRetriever,
-			0,
-		)) as BaseRetriever;
+		const model = await this.aiRootNodeContext.getModel();
+		const retriever = await this.aiRootNodeContext.getRetriever();
 
 		const items = this.getInputData();
 

diff --git a/...ages/@n8n/nodes-langchain/nodes/chains/ChainSummarization/V1/ChainSummarizationV1.node.ts b/...ages/@n8n/nodes-langchain/nodes/chains/ChainSummarization/V1/ChainSummarizationV1.node.ts
@@ -1,5 +1,4 @@
 import type { Document } from '@langchain/core/documents';
-import type { BaseLanguageModel } from '@langchain/core/language_models/base';
 import { PromptTemplate } from '@langchain/core/prompts';
 import type { SummarizationChainParams } from 'langchain/chains';
 import { loadSummarizationChain } from 'langchain/chains';
@@ -166,14 +165,11 @@ export class ChainSummarizationV1 implements INodeType {
 		this.logger.debug('Executing Vector Store QA Chain');
 		const type = this.getNodeParameter('type', 0) as 'map_reduce' | 'stuff' | 'refine';
 
-		const model = (await this.getInputConnectionData(
-			NodeConnectionType.AiLanguageModel,
-			0,
-		)) as BaseLanguageModel;
+		const model = await this.aiRootNodeContext.getModel();
 
-		const documentInput = (await this.getInputConnectionData(NodeConnectionType.AiDocument, 0)) as
-			| N8nJsonLoader
-			| Array<Document<Record<string, unknown>>>;
+		const documentInput = await this.aiRootNodeContext.getDocument<
+			N8nJsonLoader | Array<Document<Record<string, unknown>>>
+		>();
 
 		const options = this.getNodeParameter('options', 0, {}) as {
 			prompt?: string;