feat: Add deep research use case (Python)

helpers/datasources.ts

@@ -42,6 +42,16 @@ export const EXAMPLE_GDPR: TemplateDataSource = {
   },
 };
 
+export const AI_REPORTS: TemplateDataSource = {
+  type: "file",
+  config: {
+    url: new URL(
+      "https://www.europarl.europa.eu/RegData/etudes/ATAG/2024/760392/EPRS_ATA(2024)760392_EN.pdf",
+    ),
+    filename: "EPRS_ATA_2024_760392_EN.pdf",
+  },
+};
+
 export function getDataSources(
   files?: string,
   exampleFile?: boolean,

package.json

@@ -43,7 +43,7 @@
     "@types/cross-spawn": "6.0.0",
     "@types/fs-extra": "11.0.4",
     "@types/node": "^20.11.7",
-    "@types/prompts": "2.0.1",
+    "@types/prompts": "2.4.2",
     "@types/tar": "6.1.5",
     "@types/validate-npm-package-name": "3.0.0",
     "async-retry": "1.3.1",

questions/simple.ts

@@ -1,5 +1,6 @@
 import prompts from "prompts";
 import {
+  AI_REPORTS,
   EXAMPLE_10K_SEC_FILES,
   EXAMPLE_FILE,
   EXAMPLE_GDPR,
@@ -17,7 +18,8 @@ type AppType =
   | "form_filling"
   | "extractor"
   | "contract_review"
-  | "data_scientist";
+  | "data_scientist"
+  | "blog";
 
 type SimpleAnswers = {
   appType: AppType;
@@ -34,22 +36,55 @@ export const askSimpleQuestions = async (
       type: "select",
       name: "appType",
       message: "What app do you want to build?",
+      hint: "🤖: Agent, 🔀: Workflow",
       choices: [
-        { title: "Agentic RAG", value: "rag" },
-        { title: "Data Scientist", value: "data_scientist" },
         {
-          title: "Financial Report Generator (using Workflows)",
+          title: "🤖 Agentic RAG",
+          value: "rag",
+          description:
+            "Build a chatbot that can answer questions based on provided documents.",
+        },
+        {
+          title: "🤖 Data Scientist",
+          value: "data_scientist",
+          description:
+            "An data scientist agent that can analyze data and generate visualizations by using a code interpreter.",
+        },
+        {
+          title: "🤖 Code Artifact Agent",
+          value: "code_artifact",
+          description:
+            "An agent that can write code, run it in a sandboxed environment, and finally show the output in the chat UI.",
+        },
+        {
+          title: "🤖 Information Extractor",
+          value: "extractor",
+          description:
+            "Extract information from provided documents and return it as a structured JSON object by defining a Pydantic model.",
+        },
+        {
+          title: "🔀 Financial Report Generator",
           value: "financial_report_agent",
+          description:
+            "Generate a financial report by analyzing the provided 10-K SEC data and use a code interpreter to create charts or conduct further analysis.",
         },
         {
-          title: "Form Filler (using Workflows)",
+          title: "🔀 Financial 10k SEC Form Filler",
           value: "form_filling",
+          description:
+            "Extract information from 10k SEC data and use it to fill out a CSV form template.",
         },
-        { title: "Code Artifact Agent", value: "code_artifact" },
-        { title: "Information Extractor", value: "extractor" },
         {
-          title: "Contract Review (using Workflows)",
+          title: "🔀 Contract Review",
           value: "contract_review",
+          description:
+            "Extract and review contracts to ensure compliance with regulations (GDPR)",
+        },
+        {
+          title: "🔀 Blog Writer",
+          value: "blog",
+          description:
+            "Write a blog post by analyzing the provided data from different perspectives and crafting a coherent blog post with citations to the data.",
         },
       ],
     },
@@ -60,7 +95,11 @@ export const askSimpleQuestions = async (
   let llamaCloudKey = args.llamaCloudKey;
   let useLlamaCloud = false;
 
-  if (appType !== "extractor" && appType !== "contract_review") {
+  if (
+    appType !== "extractor" &&
+    appType !== "contract_review" &&
+    appType !== "blog"
+  ) {
     const { language: newLanguage } = await prompts(
       {
         type: "select",
@@ -188,6 +227,13 @@ const convertAnswers = async (
       frontend: false,
       dataSources: [EXAMPLE_GDPR],
     },
+    blog: {
+      template: "multiagent",
+      useCase: "blog",
+      tools: [],
+      frontend: true,
+      dataSources: [AI_REPORTS],
+    },
   };
   const results = lookup[answers.appType];
   return {

templates/components/agents/python/blog/README-template.md

@@ -1,16 +1,4 @@
-## Overview
-
-This example is using three agents to generate a blog post:
-
-- a researcher that retrieves content via a RAG pipeline,
-- a writer that specializes in writing blog posts and
-- a reviewer that is reviewing the blog post.
-
-There are three different methods how the agents can interact to reach their goal:
-
-1. [Choreography](./app/agents/choreography.py) - the agents decide themselves to delegate a task to another agent
-1. [Orchestrator](./app/agents/orchestrator.py) - a central orchestrator decides which agent should execute a task
-1. [Explicit Workflow](./app/agents/workflow.py) - a pre-defined workflow specific for the task is used to execute the tasks
+This is a [LlamaIndex](https://www.llamaindex.ai/) multi-agents project using [Workflows](https://docs.llamaindex.ai/en/stable/understanding/workflows/).
 
 ## Getting Started
 
@@ -35,25 +23,14 @@ Third, run the development server:
 poetry run dev
 ```
 
-Per default, the example is using the explicit workflow. You can change the example by setting the `EXAMPLE_TYPE` environment variable to `choreography` or `orchestrator`.
-The example provides one streaming API endpoint `/api/chat`.
-You can test the endpoint with the following curl request:
-
-```
-curl --location 'localhost:8000/api/chat' \
---header 'Content-Type: application/json' \
---data '{ "messages": [{ "role": "user", "content": "Write a blog post about physical standards for letters" }] }'
-```
-
-You can start editing the API by modifying `app/api/routers/chat.py` or `app/examples/workflow.py`. The API auto-updates as you save the files.
+## Use Case: Blog writer
 
-Open [http://localhost:8000](http://localhost:8000) with your browser to start the app.
+The workflow writes blog posts based on documents in the [data](./data) directory. You can start with the included PDF about AI investment in 2024, or add your own documents and run generate script again.
 
-To start the app optimized for **production**, run:
+After starting the server, go to [http://localhost:8000](http://localhost:8000) and send a message to the agent to write a blog post.
+E.g: "Write a post about AI investment in 2024"
 
-```
-poetry run prod
-```
+To update the workflow, you can edit the [writer.py](./app/workflows/writer.py) file.
 
 ## Deployments