langwatch 0.3.0-prerelease.1 → 0.3.0-prerelease.2

package/examples/langgraph/package.json

@@ -1,28 +0,0 @@
-{
-  "name": "langgraph-sdk-example",
-  "description": "",
-  "main": "index.js",
-  "scripts": {
-    "start": "tsc && dotenv -- node dist/index.js",
-    "build": "tsc",
-    "test": "echo \"Error: no test specified\" && exit 1"
-  },
-  "keywords": [],
-  "author": "LangWatch",
-  "type": "module",
-  "license": "MIT",
-  "dependencies": {
-    "@langchain/openai": "^0.6.3",
-    "@langchain/core": "^0.3.66",
-    "langchain": "^0.3.30",
-    "@langchain/langgraph": "^0.4.2",
-    "@opentelemetry/context-async-hooks": "^2.0.1",
-    "@opentelemetry/sdk-node": "^0.203.0",
-    "cli-markdown": "^3.5.1",
-    "langwatch": "file:../../"
-  },
-  "devDependencies": {
-    "typescript": "^5.9.2",
-    "dotenv-cli": "^10.0.0"
-  }
-}

package/examples/langgraph/src/cli-markdown.d.ts

@@ -1,137 +0,0 @@
-declare module 'cli-markdown' {
-  export interface CliMarkdownOptions {
-    /**
-     * Custom styles for different markdown elements
-     */
-    styles?: {
-      /**
-       * Style for headings (h1, h2, h3, etc.)
-       */
-      heading?: string;
-      /**
-       * Style for bold text
-       */
-      bold?: string;
-      /**
-       * Style for italic text
-       */
-      italic?: string;
-      /**
-       * Style for code blocks
-       */
-      code?: string;
-      /**
-       * Style for inline code
-       */
-      inlineCode?: string;
-      /**
-       * Style for links
-       */
-      link?: string;
-      /**
-       * Style for lists
-       */
-      list?: string;
-      /**
-       * Style for blockquotes
-       */
-      blockquote?: string;
-      /**
-       * Style for horizontal rules
-       */
-      hr?: string;
-    };
-    /**
-     * Whether to preserve line breaks
-     */
-    preserveLineBreaks?: boolean;
-    /**
-     * Custom renderer functions for different markdown elements
-     */
-    renderers?: {
-      /**
-       * Custom renderer for headings
-       */
-      heading?: (text: string, level: number) => string;
-      /**
-       * Custom renderer for paragraphs
-       */
-      paragraph?: (text: string) => string;
-      /**
-       * Custom renderer for lists
-       */
-      list?: (items: string[], ordered: boolean) => string;
-      /**
-       * Custom renderer for list items
-       */
-      listItem?: (text: string, index: number) => string;
-      /**
-       * Custom renderer for code blocks
-       */
-      code?: (code: string, language?: string) => string;
-      /**
-       * Custom renderer for inline code
-       */
-      inlineCode?: (code: string) => string;
-      /**
-       * Custom renderer for links
-       */
-      link?: (text: string, url: string) => string;
-      /**
-       * Custom renderer for images
-       */
-      image?: (alt: string, url: string) => string;
-      /**
-       * Custom renderer for blockquotes
-       */
-      blockquote?: (text: string) => string;
-      /**
-       * Custom renderer for horizontal rules
-       */
-      hr?: () => string;
-      /**
-       * Custom renderer for emphasis (bold/italic)
-       */
-      emphasis?: (text: string, type: 'bold' | 'italic') => string;
-    };
-    /**
-     * Maximum width for text wrapping
-     */
-    maxWidth?: number;
-    /**
-     * Whether to enable ANSI color codes
-     */
-    colors?: boolean;
-    /**
-     * Custom color theme
-     */
-    theme?: {
-      /**
-       * Color for headings
-       */
-      heading?: string;
-      /**
-       * Color for links
-       */
-      link?: string;
-      /**
-       * Color for code
-       */
-      code?: string;
-      /**
-       * Color for blockquotes
-       */
-      blockquote?: string;
-    };
-  }
-
-  /**
-   * Converts markdown text to formatted CLI output
-   * @param markdown - The markdown text to convert
-   * @param options - Configuration options for the conversion
-   * @returns Formatted CLI text
-   */
-  function cliMarkdown(markdown: string, options?: CliMarkdownOptions): string;
-
-  export = cliMarkdown;
-}

package/examples/langgraph/src/index.ts

@@ -1,196 +0,0 @@
-import { setupLangWatch } from "langwatch/node";
-import { LangWatchCallbackHandler } from "langwatch/observability/instrumentation/langchain";
-import { getLangWatchTracer } from "langwatch";
-import { semconv } from "langwatch/observability";
-import { ChatOpenAI } from "@langchain/openai";
-import { HumanMessage, SystemMessage } from "@langchain/core/messages";
-import * as readline from "readline";
-import cliMarkdown from "cli-markdown";
-
-await setupLangWatch();
-
-const tracer = getLangWatchTracer("langgraph-sdk-example");
-
-// Define the state interface for our multi-agent workflow
-interface WorkflowState {
-  userQuery: string;
-  analysis: string;
-  recommendation: string;
-  finalResponse: string;
-}
-
-async function main() {
-  const threadId = crypto.randomUUID();
-  const langWatchCallback = new LangWatchCallbackHandler();
-
-  const rl = readline.createInterface({
-    input: process.stdin,
-    output: process.stdout,
-  });
-
-  console.log('🤖 Multi-Agent Workflow started! Type "quit" to exit.');
-  console.log("This example demonstrates a workflow with:");
-  console.log("1. Query Analysis Agent");
-  console.log("2. Recommendation Agent");
-  console.log("3. Response Synthesis Agent");
-  console.log("---");
-
-  // Initialize LangChain chat model
-  const chatModel = new ChatOpenAI({
-    modelName: "gpt-4o-mini",
-    temperature: 0.7,
-    callbacks: [langWatchCallback],
-  });
-
-  // Agent 1: Query Analysis Agent
-  const analyzeQuery = async (userQuery: string): Promise<string> => {
-    const analysisPrompt = new SystemMessage(
-      "You are an expert at analyzing user queries. Analyze the user's query and provide insights about what they're asking for, their intent, and any specific requirements."
-    );
-
-    const result = await chatModel.invoke([
-      analysisPrompt,
-      new HumanMessage(`Analyze this query: ${userQuery}`)
-    ]);
-
-    return result.content as string;
-  };
-
-  // Agent 2: Recommendation Agent
-  const generateRecommendation = async (userQuery: string, analysis: string): Promise<string> => {
-    const recommendationPrompt = new SystemMessage(
-      "You are an expert at providing helpful recommendations. Based on the analysis, provide specific, actionable recommendations."
-    );
-
-    const result = await chatModel.invoke([
-      recommendationPrompt,
-      new HumanMessage(`Based on this analysis: ${analysis}\n\nProvide recommendations for: ${userQuery}`)
-    ]);
-
-    return result.content as string;
-  };
-
-  // Agent 3: Response Synthesis Agent
-  const synthesizeResponse = async (userQuery: string, analysis: string, recommendation: string): Promise<string> => {
-    const synthesisPrompt = new SystemMessage(
-      "You are an expert at synthesizing information into clear, helpful responses. Combine the analysis and recommendations into a comprehensive, well-structured response."
-    );
-
-    const result = await chatModel.invoke([
-      synthesisPrompt,
-      new HumanMessage(`Original Query: ${userQuery}\n\nAnalysis: ${analysis}\n\nRecommendations: ${recommendation}\n\nSynthesize this into a comprehensive response.`)
-    ]);
-
-    return result.content as string;
-  };
-
-  // Main workflow function
-  const runWorkflow = async (userQuery: string): Promise<WorkflowState> => {
-    console.log("🔍 Step 1: Analyzing query...");
-    const analysis = await analyzeQuery(userQuery);
-
-    console.log("💡 Step 2: Generating recommendations...");
-    const recommendation = await generateRecommendation(userQuery, analysis);
-
-    console.log("🎯 Step 3: Synthesizing final response...");
-    const finalResponse = await synthesizeResponse(userQuery, analysis, recommendation);
-
-    return {
-      userQuery,
-      analysis,
-      recommendation,
-      finalResponse,
-    };
-  };
-
-  while (true) {
-    let finish = false;
-
-    await tracer.withActiveSpan("workflow_iteration", {
-      attributes: {
-        [semconv.ATTR_LANGWATCH_THREAD_ID]: threadId,
-      },
-    }, async (span) => {
-      try {
-        // Get user input
-        const userInput = await new Promise<string>((resolve) => {
-          rl.question("You: ", resolve);
-        });
-
-        // Check for exit command
-        if (
-          userInput.toLowerCase() === "quit" ||
-          userInput.toLowerCase() === "exit"
-        ) {
-          console.log("👋 Goodbye!");
-          finish = true;
-          return;
-        }
-
-        // Skip empty input
-        if (!userInput.trim()) {
-          return;
-        }
-
-        console.log("🤖 Processing through multi-agent workflow...");
-
-        // Execute the workflow
-        const result = await runWorkflow(userInput);
-
-        // Display the workflow results
-        console.log("\n📊 Workflow Results:");
-        console.log("---");
-
-        console.log("🔍 Analysis:");
-        console.log(cliMarkdown(result.analysis, {
-          colors: true,
-          maxWidth: 80,
-          theme: {
-            heading: "cyan",
-            link: "blue",
-            code: "green",
-            blockquote: "yellow",
-          },
-        }));
-
-        console.log("\n💡 Recommendations:");
-        console.log(cliMarkdown(result.recommendation, {
-          colors: true,
-          maxWidth: 80,
-          theme: {
-            heading: "cyan",
-            link: "blue",
-            code: "green",
-            blockquote: "yellow",
-          },
-        }));
-
-        console.log("\n🎯 Final Response:");
-        console.log(cliMarkdown(result.finalResponse, {
-          colors: true,
-          maxWidth: 80,
-          theme: {
-            heading: "cyan",
-            link: "blue",
-            code: "green",
-            blockquote: "yellow",
-          },
-        }));
-
-        console.log("---");
-      } catch (error) {
-        console.error("❌ Error:", error);
-        console.log("Please try again.");
-      }
-    });
-
-    if (finish) {
-      break;
-    }
-  }
-
-  rl.close();
-}
-
-// Run the multi-agent workflow
-main().catch(console.error);

package/examples/langgraph/tsconfig.json

@@ -1,25 +0,0 @@
-{
-  "compilerOptions": {
-    "target": "ES2022",
-    "module": "node16",
-    "moduleResolution": "node16",
-    "outDir": "./dist",
-    "rootDir": "./src",
-    "strict": true,
-    "esModuleInterop": true,
-    "skipLibCheck": true,
-    "forceConsistentCasingInFileNames": true,
-    "allowSyntheticDefaultImports": true,
-    "resolveJsonModule": true,
-    "declaration": true,
-    "declarationMap": true,
-    "sourceMap": true
-  },
-  "include": [
-    "src/**/*"
-  ],
-  "exclude": [
-    "node_modules",
-    "dist"
-  ]
-}

package/examples/mastra/.env.example

@@ -1,2 +0,0 @@
-LANGWATCH_API_KEY=""
-OPENAI_API_KEY=""

package/examples/mastra/README.md

@@ -1,57 +0,0 @@
-# Mastra Weather Agent Example
-
-This example demonstrates how to use LangWatch with Mastra to create a weather agent that can provide weather information and suggest activities based on weather conditions.
-
-## Features
-
-- **Weather Agent**: An AI agent that can fetch weather data and suggest activities
-- **Weather Tool**: A tool that fetches real-time weather data from Open-Meteo API
-- **CLI Chatbox Interface**: Interactive command-line interface for chatting with the weather agent
-- **LangWatch Integration**: Full observability and tracing with LangWatch
-
-## Getting Started
-
-1. Install dependencies:
-   ```bash
-   npm install
-   ```
-
-2. Set up your environment variables:
-   ```bash
-   cp .env.example .env
-   # Add your OpenAI API key to .env
-   ```
-
-3. Run the CLI chatbox interface:
-   ```bash
-   npm run cli
-   ```
-
-4. Or run the workflow example:
-   ```bash
-   npm start
-   ```
-
-## Usage
-
-### CLI Chatbox Interface
-
-The CLI interface allows you to interact with the weather agent in a conversational manner:
-
-```bash
-npm run cli
-```
-
-Example conversation:
-```
-🌤️  Weather Agent Chatbot started! Type "quit" to exit.
-Ask me about weather for any location and I'll help you plan activities!
----
-You: What's the weather like in Paris?
-🌤️  Checking weather and planning activities...
-[Agent responds with weather information and activity suggestions]
-```
-
-### Workflow Example
-
-The workflow example demonstrates how to use Mastra workflows to fetch weather data and plan activities programmatically.