add data scientist use case (directly using uploaded files) (#355)

---------
Co-authored-by: Marcus Schiesser <[email protected]>
Co-authored-by: Thuc Pham <[email protected]>

Author: 3 people

.changeset/poor-knives-smoke.md

@@ -0,0 +1,5 @@
+---
+"create-llama": patch
+---
+
+Fix event streaming is blocked

.changeset/wet-tips-judge.md

@@ -0,0 +1,5 @@
+---
+"create-llama": patch
+---
+
+Add upload file to sandbox (artifact and code interpreter)

helpers/tools.ts

@@ -139,7 +139,7 @@ For better results, you can specify the region parameter to get results from a s
     dependencies: [
       {
         name: "e2b_code_interpreter",
-        version: "0.0.10",
+        version: "0.0.11b38",
       },
     ],
     supportedFrameworks: ["fastapi", "express", "nextjs"],

questions/simple.ts

@@ -5,14 +5,19 @@ import { getTools } from "../helpers/tools";
 import { ModelConfig, TemplateFramework } from "../helpers/types";
 import { PureQuestionArgs, QuestionResults } from "./types";
 import { askPostInstallAction, questionHandlers } from "./utils";
-type AppType = "rag" | "code_artifact" | "multiagent" | "extractor";
+
+type AppType =
+  | "rag"
+  | "code_artifact"
+  | "multiagent"
+  | "extractor"
+  | "data_scientist";
 
 type SimpleAnswers = {
   appType: AppType;
   language: TemplateFramework;
   useLlamaCloud: boolean;
   llamaCloudKey?: string;
-  modelConfig: ModelConfig;
 };
 
 export const askSimpleQuestions = async (
@@ -25,6 +30,7 @@ export const askSimpleQuestions = async (
       message: "What app do you want to build?",
       choices: [
         { title: "Agentic RAG", value: "rag" },
+        { title: "Data Scientist", value: "data_scientist" },
         { title: "Code Artifact Agent", value: "code_artifact" },
         { title: "Multi-Agent Report Gen", value: "multiagent" },
         { title: "Structured extraction", value: "extractor" },
@@ -80,40 +86,56 @@ export const askSimpleQuestions = async (
     }
   }
 
-  const modelConfig = await askModelConfig({
-    openAiKey: args.openAiKey,
-    askModels: args.askModels ?? false,
-    framework: language,
-  });
-
-  const results = convertAnswers({
+  const results = await convertAnswers(args, {
     appType,
     language,
     useLlamaCloud,
     llamaCloudKey,
-    modelConfig,
   });
 
   results.postInstallAction = await askPostInstallAction(results);
   return results;
 };
 
-const convertAnswers = (answers: SimpleAnswers): QuestionResults => {
+const convertAnswers = async (
+  args: PureQuestionArgs,
+  answers: SimpleAnswers,
+): Promise<QuestionResults> => {
+  const MODEL_GPT4o: ModelConfig = {
+    provider: "openai",
+    apiKey: args.openAiKey,
+    model: "gpt-4o",
+    embeddingModel: "text-embedding-3-large",
+    dimensions: 1536,
+    isConfigured(): boolean {
+      return !!args.openAiKey;
+    },
+  };
   const lookup: Record<
     AppType,
-    Pick<QuestionResults, "template" | "tools" | "frontend" | "dataSources">
+    Pick<QuestionResults, "template" | "tools" | "frontend" | "dataSources"> & {
+      modelConfig?: ModelConfig;
+    }
   > = {
     rag: {
       template: "streaming",
       tools: getTools(["duckduckgo"]),
       frontend: true,
       dataSources: [EXAMPLE_FILE],
     },
+    data_scientist: {
+      template: "streaming",
+      tools: getTools(["interpreter", "document_generator"]),
+      frontend: true,
+      dataSources: [],
+      modelConfig: MODEL_GPT4o,
+    },
     code_artifact: {
       template: "streaming",
       tools: getTools(["artifact"]),
       frontend: true,
       dataSources: [],
+      modelConfig: MODEL_GPT4o,
     },
     multiagent: {
       template: "multiagent",
@@ -140,11 +162,16 @@ const convertAnswers = (answers: SimpleAnswers): QuestionResults => {
     llamaCloudKey: answers.llamaCloudKey,
     useLlamaParse: answers.useLlamaCloud,
     llamapack: "",
-    postInstallAction: "none",
     vectorDb: answers.useLlamaCloud ? "llamacloud" : "none",
-    modelConfig: answers.modelConfig,
     observability: "none",
     ...results,
+    modelConfig:
+      results.modelConfig ??
+      (await askModelConfig({
+        openAiKey: args.openAiKey,
+        askModels: args.askModels ?? false,
+        framework: answers.language,
+      })),
     frontend: answers.language === "nextjs" ? false : results.frontend,
   };
 };

templates/components/engines/python/agent/tools/artifact.py

@@ -66,21 +66,29 @@ class CodeGeneratorTool:
     def __init__(self):
         pass
 
-    def artifact(self, query: str, old_code: Optional[str] = None) -> Dict:
-        """Generate a code artifact based on the input.
+    def artifact(
+        self,
+        query: str,
+        sandbox_files: Optional[List[str]] = None,
+        old_code: Optional[str] = None,
+    ) -> Dict:
+        """Generate a code artifact based on the provided input.
 
         Args:
-            query (str): The description of the application you want to build.
+            query (str): A description of the application you want to build.
+            sandbox_files (Optional[List[str]], optional): A list of sandbox file paths. Defaults to None. Include these files if the code requires them.
             old_code (Optional[str], optional): The existing code to be modified. Defaults to None.
 
         Returns:
-            Dict: A dictionary containing the generated artifact information.
+            Dict: A dictionary containing information about the generated artifact.
         """
 
         if old_code:
             user_message = f"{query}\n\nThe existing code is: \n```\n{old_code}\n```"
         else:
             user_message = query
+        if sandbox_files:
+            user_message += f"\n\nThe provided files are: \n{str(sandbox_files)}"
 
         messages: List[ChatMessage] = [
             ChatMessage(role="system", content=CODE_GENERATION_PROMPT),
@@ -90,7 +98,10 @@ def artifact(self, query: str, old_code: Optional[str] = None) -> Dict:
             sllm = Settings.llm.as_structured_llm(output_cls=CodeArtifact)  # type: ignore
             response = sllm.chat(messages)
             data: CodeArtifact = response.raw
-            return data.model_dump()
+            data_dict = data.model_dump()
+            if sandbox_files:
+                data_dict["files"] = sandbox_files
+            return data_dict
         except Exception as e:
             logger.error(f"Failed to generate artifact: {str(e)}")
             raise e