@librechat/agents 2.4.322 → 3.0.0-rc10

package/src/scripts/multi-agent-sequence.ts

@@ -0,0 +1,212 @@
+import { config } from 'dotenv';
+config();
+
+import { HumanMessage, BaseMessage } from '@langchain/core/messages';
+import { Run } from '@/run';
+import { Providers, GraphEvents } from '@/common';
+import { ChatModelStreamHandler, createContentAggregator } from '@/stream';
+import { ToolEndHandler, ModelEndHandler } from '@/events';
+import type * as t from '@/types';
+
+const conversationHistory: BaseMessage[] = [];
+
+/**
+ * Example of simple sequential multi-agent system
+ *
+ * Graph structure:
+ * START -> agent_a -> agent_b -> agent_c -> END
+ *
+ * No conditions, no tools, just automatic sequential flow
+ */
+async function testSequentialMultiAgent() {
+  console.log('Testing Sequential Multi-Agent System (A → B → C)...\n');
+
+  // Set up content aggregator
+  const { contentParts, aggregateContent } = createContentAggregator();
+
+  // Define three simple agents
+  const agents: t.AgentInputs[] = [
+    {
+      agentId: 'agent_a',
+      provider: Providers.ANTHROPIC,
+      clientOptions: {
+        modelName: 'claude-3-5-sonnet-latest',
+        apiKey: process.env.ANTHROPIC_API_KEY,
+      },
+      instructions: `You are Agent A, the first agent in a sequential workflow.
+      Your job is to:
+      1. Receive the initial user request
+      2. Process it and add your perspective (keep it brief, 2-3 sentences)
+      3. Pass it along to Agent B
+      
+      Start your response with "AGENT A:" and end with "Passing to Agent B..."`,
+      maxContextTokens: 8000,
+    },
+    {
+      agentId: 'agent_b',
+      provider: Providers.ANTHROPIC,
+      clientOptions: {
+        modelName: 'claude-3-5-sonnet-latest',
+        apiKey: process.env.ANTHROPIC_API_KEY,
+      },
+      instructions: `You are Agent B, the second agent in a sequential workflow.
+      Your job is to:
+      1. Receive the context from Agent A
+      2. Add your own analysis or perspective (keep it brief, 2-3 sentences)
+      3. Pass it along to Agent C
+      
+      Start your response with "AGENT B:" and end with "Passing to Agent C..."`,
+      maxContextTokens: 8000,
+    },
+    {
+      agentId: 'agent_c',
+      provider: Providers.ANTHROPIC,
+      clientOptions: {
+        modelName: 'claude-3-5-sonnet-latest',
+        apiKey: process.env.ANTHROPIC_API_KEY,
+      },
+      instructions: `You are Agent C, the final agent in a sequential workflow.
+      Your job is to:
+      1. Receive the context from Agents A and B
+      2. Provide a final summary or conclusion based on all previous inputs
+      3. Complete the workflow
+      
+      Start your response with "AGENT C:" and end with "Workflow complete."`,
+      maxContextTokens: 8000,
+    },
+  ];
+
+  // Define sequential edges using 'direct' type to avoid tool creation
+  // This creates automatic transitions without requiring agents to call tools
+  const edges: t.GraphEdge[] = [
+    {
+      from: 'agent_a',
+      to: 'agent_b',
+      edgeType: 'direct', // This creates direct edges without tools
+      description: 'Automatic transition from A to B',
+    },
+    {
+      from: 'agent_b',
+      to: 'agent_c',
+      edgeType: 'direct', // This creates direct edges without tools
+      description: 'Automatic transition from B to C',
+    },
+  ];
+
+  // Track agent progression
+  let currentAgent = '';
+
+  // Create custom handlers
+  const customHandlers = {
+    [GraphEvents.TOOL_END]: new ToolEndHandler(),
+    [GraphEvents.CHAT_MODEL_END]: new ModelEndHandler(),
+    [GraphEvents.CHAT_MODEL_STREAM]: new ChatModelStreamHandler(),
+    [GraphEvents.ON_RUN_STEP]: {
+      handle: (
+        event: GraphEvents.ON_RUN_STEP,
+        data: t.StreamEventData
+      ): void => {
+        const runStepData = data as any;
+        if (runStepData?.name) {
+          currentAgent = runStepData.name;
+          console.log(`\n→ ${currentAgent} is processing...`);
+        }
+        aggregateContent({ event, data: data as t.RunStep });
+      },
+    },
+    [GraphEvents.ON_RUN_STEP_COMPLETED]: {
+      handle: (
+        event: GraphEvents.ON_RUN_STEP_COMPLETED,
+        data: t.StreamEventData
+      ): void => {
+        const runStepData = data as any;
+        if (runStepData?.name) {
+          console.log(`✓ ${runStepData.name} completed`);
+        }
+        aggregateContent({
+          event,
+          data: data as unknown as { result: t.ToolEndEvent },
+        });
+      },
+    },
+    [GraphEvents.ON_RUN_STEP_DELTA]: {
+      handle: (
+        event: GraphEvents.ON_RUN_STEP_DELTA,
+        data: t.StreamEventData
+      ): void => {
+        aggregateContent({ event, data: data as t.RunStepDeltaEvent });
+      },
+    },
+    [GraphEvents.ON_MESSAGE_DELTA]: {
+      handle: (
+        event: GraphEvents.ON_MESSAGE_DELTA,
+        data: t.StreamEventData
+      ): void => {
+        aggregateContent({ event, data: data as t.MessageDeltaEvent });
+      },
+    },
+  };
+
+  // Create multi-agent run configuration
+  const runConfig: t.RunConfig = {
+    runId: `sequential-multi-agent-${Date.now()}`,
+    graphConfig: {
+      type: 'multi-agent',
+      agents,
+      edges,
+    },
+    customHandlers,
+    returnContent: true,
+  };
+
+  try {
+    // Create and execute the run
+    const run = await Run.create(runConfig);
+
+    // Test with a simple question
+    const userMessage =
+      'What are the key considerations for building a recommendation system?';
+    conversationHistory.push(new HumanMessage(userMessage));
+
+    console.log(`User: "${userMessage}"\n`);
+    console.log('Starting sequential workflow...\n');
+
+    const config = {
+      configurable: {
+        thread_id: 'sequential-conversation-1',
+      },
+      streamMode: 'values',
+      version: 'v2' as const,
+    };
+
+    // Process with streaming
+    const inputs = {
+      messages: conversationHistory,
+    };
+
+    const finalContentParts = await run.processStream(inputs, config);
+    const finalMessages = run.getRunMessages();
+
+    if (finalMessages) {
+      conversationHistory.push(...finalMessages);
+    }
+
+    console.log('\n\n=== Final Output ===');
+    console.log('Sequential flow completed successfully!');
+    console.log(`Total content parts: ${contentParts.length}`);
+
+    // Display the sequential responses
+    const aiMessages = conversationHistory.filter(
+      (msg) => msg._getType() === 'ai'
+    );
+    aiMessages.forEach((msg, index) => {
+      console.log(`\n--- Response ${index + 1} ---`);
+      console.log(msg.content);
+    });
+  } catch (error) {
+    console.error('Error in sequential multi-agent test:', error);
+  }
+}
+
+// Run the test
+testSequentialMultiAgent();

package/src/scripts/multi-agent-supervisor.ts

@@ -0,0 +1,362 @@
+import { config } from 'dotenv';
+config();
+
+import { HumanMessage, BaseMessage } from '@langchain/core/messages';
+import { Run } from '@/run';
+import { Providers, GraphEvents } from '@/common';
+import { ChatModelStreamHandler, createContentAggregator } from '@/stream';
+import { ToolEndHandler, ModelEndHandler } from '@/events';
+import type * as t from '@/types';
+
+const conversationHistory: BaseMessage[] = [];
+
+/**
+ * Example of supervisor-based multi-agent system
+ *
+ * The supervisor has handoff tools for 5 different specialists.
+ * To demonstrate the concept while respecting LangGraph constraints,
+ * we show two approaches:
+ *
+ * 1. All 5 specialists exist but share the same adaptive configuration
+ * 2. Only 2 agents total by using a single adaptive specialist (requires workaround)
+ */
+async function testSupervisorMultiAgent() {
+  console.log('Testing Supervisor-Based Multi-Agent System...\n');
+
+  // NOTE: To truly have only 2 agents with 5 handoff tools, you would need:
+  // 1. Custom tool implementation (see multi-agent-supervisor-mock.ts)
+  // 2. Or modify MultiAgentGraph to support "virtual" agents
+  // 3. Or use a single conditional edge with role parameter
+  //
+  // This example shows the concept using 6 agents that share configuration
+
+  // Set up content aggregator
+  const { contentParts, aggregateContent } = createContentAggregator();
+
+  // Define configurations for all possible specialists
+  const specialistConfigs = {
+    data_analyst: {
+      provider: Providers.OPENAI,
+      clientOptions: {
+        modelName: 'gpt-4.1',
+        apiKey: process.env.OPENAI_API_KEY,
+      },
+      instructions: `You are a Data Analyst specialist. Your expertise includes:
+      - Statistical analysis and data visualization
+      - SQL queries and database optimization
+      - Python/R for data science
+      - Machine learning model evaluation
+      - A/B testing and experiment design
+      
+      Follow the supervisor's specific instructions carefully.`,
+      maxContextTokens: 8000,
+    },
+    security_expert: {
+      provider: Providers.OPENAI,
+      clientOptions: {
+        modelName: 'gpt-4.1',
+        apiKey: process.env.OPENAI_API_KEY,
+      },
+      instructions: `You are a Security Expert. Your expertise includes:
+      - Cybersecurity best practices
+      - Vulnerability assessment and penetration testing
+      - Security architecture and threat modeling
+      - Compliance (GDPR, HIPAA, SOC2, etc.)
+      - Incident response and forensics
+      
+      Follow the supervisor's specific instructions carefully.`,
+      maxContextTokens: 8000,
+    },
+    product_designer: {
+      provider: Providers.OPENAI,
+      clientOptions: {
+        modelName: 'gpt-4.1',
+        apiKey: process.env.OPENAI_API_KEY,
+      },
+      instructions: `You are a Product Designer. Your expertise includes:
+      - User experience (UX) design principles
+      - User interface (UI) design and prototyping
+      - Design systems and component libraries
+      - User research and usability testing
+      - Accessibility and inclusive design
+      
+      Follow the supervisor's specific instructions carefully.`,
+      maxContextTokens: 8000,
+    },
+    devops_engineer: {
+      provider: Providers.OPENAI,
+      clientOptions: {
+        modelName: 'gpt-4.1',
+        apiKey: process.env.OPENAI_API_KEY,
+      },
+      instructions: `You are a DevOps Engineer. Your expertise includes:
+      - CI/CD pipeline design and optimization
+      - Infrastructure as Code (Terraform, CloudFormation)
+      - Container orchestration (Kubernetes, Docker)
+      - Cloud platforms (AWS, GCP, Azure)
+      - Monitoring, logging, and observability
+      
+      Follow the supervisor's specific instructions carefully.`,
+      maxContextTokens: 8000,
+    },
+    legal_advisor: {
+      provider: Providers.OPENAI,
+      clientOptions: {
+        modelName: 'gpt-4.1',
+        apiKey: process.env.OPENAI_API_KEY,
+      },
+      instructions: `You are a Legal Advisor specializing in technology. Your expertise includes:
+      - Software licensing and open source compliance
+      - Data privacy and protection laws
+      - Intellectual property and patents
+      - Contract review and negotiation
+      - Regulatory compliance for tech companies
+      
+      Follow the supervisor's specific instructions carefully.`,
+      maxContextTokens: 8000,
+    },
+  };
+
+  // Track which specialist role was selected
+  let selectedRole = '';
+  let roleInstructions = '';
+
+  // Create custom handlers
+  const customHandlers = {
+    [GraphEvents.TOOL_END]: new ToolEndHandler(),
+    [GraphEvents.CHAT_MODEL_END]: new ModelEndHandler(),
+    [GraphEvents.CHAT_MODEL_STREAM]: new ChatModelStreamHandler(),
+    [GraphEvents.ON_RUN_STEP]: {
+      handle: (
+        event: GraphEvents.ON_RUN_STEP,
+        data: t.StreamEventData
+      ): void => {
+        const runStepData = data as any;
+        if (runStepData?.name) {
+          console.log(`\n[${runStepData.name}] Processing...`);
+        }
+        aggregateContent({ event, data: data as t.RunStep });
+      },
+    },
+    [GraphEvents.ON_RUN_STEP_COMPLETED]: {
+      handle: (
+        event: GraphEvents.ON_RUN_STEP_COMPLETED,
+        data: t.StreamEventData
+      ): void => {
+        aggregateContent({
+          event,
+          data: data as unknown as { result: t.ToolEndEvent },
+        });
+      },
+    },
+    [GraphEvents.ON_MESSAGE_DELTA]: {
+      handle: (
+        event: GraphEvents.ON_MESSAGE_DELTA,
+        data: t.StreamEventData
+      ): void => {
+        console.dir(data, { depth: null });
+        aggregateContent({ event, data: data as t.MessageDeltaEvent });
+      },
+    },
+    [GraphEvents.TOOL_START]: {
+      handle: (
+        _event: string,
+        data: t.StreamEventData,
+        metadata?: Record<string, unknown>
+      ): void => {
+        const toolData = data as any;
+        if (toolData?.name?.includes('transfer_to_')) {
+          const specialist = toolData.name.replace('transfer_to_', '');
+          console.log(`\n🔀 Transferring to ${specialist}...`);
+          selectedRole = specialist;
+        }
+      },
+    },
+  };
+
+  // Function to create the graph with supervisor having multiple handoff options
+  function createSupervisorGraph(): t.RunConfig {
+    console.log(`\nCreating graph with supervisor and 5 specialist agents.`);
+    console.log('All specialists share the same adaptive configuration.\n');
+
+    // Define the adaptive specialist configuration that will be reused
+    const specialistConfig = {
+      provider: Providers.OPENAI,
+      clientOptions: {
+        modelName: 'gpt-4.1',
+        apiKey: process.env.OPENAI_API_KEY,
+      },
+      instructions: `You are an Adaptive Specialist. Your agent ID indicates your role:
+      
+      - data_analyst: Focus on statistical analysis, metrics, ML evaluation, A/B testing
+      - security_expert: Focus on cybersecurity, vulnerability assessment, compliance  
+      - product_designer: Focus on UX/UI design, user research, accessibility
+      - devops_engineer: Focus on CI/CD, infrastructure, cloud platforms, monitoring
+      - legal_advisor: Focus on licensing, privacy laws, contracts, regulatory compliance
+      
+      The supervisor will provide specific instructions. Follow them while maintaining your expert perspective.`,
+      maxContextTokens: 8000,
+    };
+
+    // Create the graph with supervisor and all 5 specialists
+    // All specialists share the same adaptive configuration
+    const agents: t.AgentInputs[] = [
+      {
+        agentId: 'supervisor',
+        provider: Providers.OPENAI,
+        clientOptions: {
+          modelName: 'gpt-4.1-mini',
+          apiKey: process.env.OPENAI_API_KEY,
+        },
+        instructions: `You are a Task Supervisor with access to 5 specialist agents:
+        1. transfer_to_data_analyst - For statistical analysis and metrics
+        2. transfer_to_security_expert - For cybersecurity and vulnerability assessment  
+        3. transfer_to_product_designer - For UX/UI design
+        4. transfer_to_devops_engineer - For infrastructure and deployment
+        5. transfer_to_legal_advisor - For compliance and licensing
+        
+        Your role is to:
+        1. Analyze the incoming request
+        2. Decide which specialist is best suited
+        3. Use the appropriate transfer tool (e.g., transfer_to_data_analyst)
+        4. Provide specific instructions to guide their work
+        
+        Be specific about what you need from the specialist.`,
+        maxContextTokens: 8000,
+      },
+      // Include all 5 specialists with the same adaptive configuration
+      {
+        agentId: 'data_analyst',
+        ...specialistConfig,
+      },
+      {
+        agentId: 'security_expert',
+        ...specialistConfig,
+      },
+      {
+        agentId: 'product_designer',
+        ...specialistConfig,
+      },
+      {
+        agentId: 'devops_engineer',
+        ...specialistConfig,
+      },
+      {
+        agentId: 'legal_advisor',
+        ...specialistConfig,
+      },
+    ];
+
+    // Create edges from supervisor to all 5 specialists
+    const edges: t.GraphEdge[] = [
+      {
+        from: 'supervisor',
+        to: 'data_analyst',
+        description:
+          'Transfer to data analyst for statistical analysis and metrics',
+        edgeType: 'handoff',
+      },
+      {
+        from: 'supervisor',
+        to: 'security_expert',
+        description: 'Transfer to security expert for cybersecurity assessment',
+        edgeType: 'handoff',
+      },
+      {
+        from: 'supervisor',
+        to: 'product_designer',
+        description: 'Transfer to product designer for UX/UI design',
+        edgeType: 'handoff',
+      },
+      {
+        from: 'supervisor',
+        to: 'devops_engineer',
+        description:
+          'Transfer to DevOps engineer for infrastructure and deployment',
+        edgeType: 'handoff',
+      },
+      {
+        from: 'supervisor',
+        to: 'legal_advisor',
+        description: 'Transfer to legal advisor for compliance and licensing',
+        edgeType: 'handoff',
+      },
+    ];
+
+    return {
+      runId: `supervisor-multi-agent-${Date.now()}`,
+      graphConfig: {
+        type: 'multi-agent',
+        agents,
+        edges,
+      },
+      customHandlers,
+      returnContent: true,
+    };
+  }
+
+  try {
+    // Test with different queries
+    const testQueries = [
+      'How can we analyze user engagement metrics to improve our product?',
+      // 'What security measures should we implement for our new API?',
+      // 'Can you help design a better onboarding flow for our mobile app?',
+      // 'We need to set up a CI/CD pipeline for our microservices.',
+      // 'What are the legal implications of using GPL-licensed code in our product?',
+    ];
+
+    const config = {
+      configurable: {
+        thread_id: 'supervisor-conversation-1',
+      },
+      streamMode: 'values',
+      version: 'v2' as const,
+    };
+
+    for (const query of testQueries) {
+      console.log(`\n${'='.repeat(60)}`);
+      console.log(`USER QUERY: "${query}"`);
+      console.log('='.repeat(60));
+
+      // Reset conversation
+      conversationHistory.length = 0;
+      conversationHistory.push(new HumanMessage(query));
+
+      // Create graph with supervisor having 5 handoff tools to 1 adaptive specialist
+      const runConfig = createSupervisorGraph();
+      const run = await Run.create(runConfig);
+
+      console.log('Processing request...');
+
+      // Process with streaming
+      const inputs = {
+        messages: conversationHistory,
+      };
+
+      const finalContentParts = await run.processStream(inputs, config);
+      const finalMessages = run.getRunMessages();
+
+      if (finalMessages) {
+        conversationHistory.push(...finalMessages);
+      }
+
+      // Show summary
+      console.log(`\n${'─'.repeat(60)}`);
+      console.log(`Agents in graph: 6 total (supervisor + 5 specialists)`);
+      console.log(`All specialists share the same adaptive configuration`);
+      console.log(
+        `Supervisor tools: transfer_to_data_analyst, transfer_to_security_expert,`
+      );
+      console.log(
+        `                 transfer_to_product_designer, transfer_to_devops_engineer,`
+      );
+      console.log(`                 transfer_to_legal_advisor`);
+      console.log('─'.repeat(60));
+    }
+  } catch (error) {
+    console.error('Error in supervisor multi-agent test:', error);
+  }
+}
+
+// Run the test
+testSupervisorMultiAgent();