fix(safe-parse): add safe JSON parse utility and apply to AI services… #49

morningstarxcdcode · 2026-01-03T17:41:03Z

This pull request sets up the initial backend project structure for LocalMind, including environment configuration, code formatting, and the foundation for AI model integrations (Google Gemini and Groq). It introduces configuration files, scripts, and the first implementation of AI-related API endpoints and services.

Project Setup & Configuration

Added .env.example with all necessary environment variables for server, database, JWT, Redis, uploads, and API keys.
Initialized package.json with project metadata, scripts for development, testing, formatting, and dependencies for Express, LangChain, AI models, and utilities.
Added Prettier config and ignore files for code formatting, and Jest configuration for testing. [1] [2] [3]

AI Model Integration: Google Gemini

Implemented GoogleService, GoogleController, and GoogleRoutes to provide an API endpoint for interacting with the Gemini AI model, using a customizable prompt template and safe JSON parsing. [1] [2] [3]
Added GeminiUtils class to configure and invoke the Google Gemini model via LangChain, supporting dynamic prompt formatting and response parsing.

AI Model Integration: Groq

Implemented GroqService, GroqController, and GroqRouter for Groq model chat endpoint, with error handling and configurable model/apiKey support. [1] [2] [3]
Defined constants and types for Groq integration to standardize responses and request payloads. [1] [2]

Prompt Template System

Added AiTemplate class and prompt text file to standardize AI assistant behavior and response format, enforcing strict JSON output and privacy-respecting instructions. [1] [2]

Developer Utilities

Added a shell script setup-cloudflare.sh to automate Cloudflared tunnel setup for local development, including installation and authentication steps.\n\nFixes 🐛 Unhandled JSON parsing in AI services and dataset processing #42 - Adds 'safeParse' utility and applies it across Google, Groq, Ollama, and DataSet services to avoid unhandled JSON parse errors.

… and dataset; add tests

gemini-code-assist · 2026-01-03T17:41:49Z

Summary of Changes

Hello @morningstarxcdcode, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request lays the groundwork for the LocalMind backend, focusing on establishing a robust and scalable architecture. It integrates multiple AI models, sets up a flexible prompt templating system, and implements essential user authentication and AI agent configuration features. The changes also include utilities for safe JSON parsing and automated development environment setup, ensuring a solid foundation for future development.

Highlights

Initial Backend Project Structure: This pull request establishes the foundational backend project structure for LocalMind, including environment configuration, code formatting with Prettier, and testing setup with Jest.
AI Model Integrations: Core integrations for Google Gemini, Groq, and Ollama AI models have been implemented, providing dedicated services, controllers, and API routes for interacting with each model.
Prompt Template System: A standardized prompt template system (AiTemplate) is introduced to ensure consistent AI assistant behavior and response formatting, enforcing strict JSON output.
User Authentication and API Key Management: User registration, login, profile management, and secure API key generation/retrieval functionalities are now in place, complete with JWT-based authentication middleware.
AI Model Configuration: A new module for configuring AI agents is added, allowing users to define and manage different AI models (cloud or on-premise) with associated keys and system prompts.
Dataset Processing: Functionality to upload and process datasets (e.g., CSV files) is introduced, leveraging LangChain's CSVLoader and Google Gemini for intelligent data extraction and validation.
Safe JSON Parsing Utility: A safeParse utility has been added to robustly handle JSON parsing, providing a fallback mechanism and error logging to prevent application crashes from malformed AI responses.
Developer Utilities: A shell script (setup-cloudflare.sh) is included to automate Cloudflared tunnel setup for local development, simplifying external access to the local server.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	`/gemini review`	Performs a code review for the current pull request in its current state.
Pull Request Summary	`/gemini summary`	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	`/gemini help`	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

gemini-code-assist

Code Review

This pull request introduces a substantial amount of new functionality, including the initial project structure, AI service integrations for Google, Groq, and Ollama, and a new safeParse utility. The setup is comprehensive, but there are several critical issues that need to be addressed. These include non-existent package versions in package.json that will break the build, required environment variables that seem to be for testing purposes and will cause the app to crash, and several logical bugs in the AI model configuration and user services. I've left detailed comments on these issues, along with suggestions for fixes. Great start, and looking forward to seeing these issues resolved!

src/api/v1/AiModelConfig/AiModelConfig.type.ts

src/validator/env.ts

package.json

src/api/v1/user/user.service.ts

src/api/v1/Ai-model/Groq/Groq.service.ts

src/api/v1/user/__test__/user.test.ts

src/api/v1/user/user.controller.ts

src/api/v1/user/user.middleware.ts

src/api/v1/Ai-model/Google/Google.service.ts

src/Template/v1/Ai.template.ts

Copilot

Pull request overview

This pull request establishes the foundational backend infrastructure for LocalMind, a platform enabling users to interact with AI models locally or via cloud APIs. The changes introduce environment configuration, user authentication, AI model integrations (Google Gemini, Groq, Ollama), a safe JSON parsing utility, dataset processing capabilities, and project tooling for development and testing.

Key Changes:

Environment validation schema with support for server, database, JWT, Redis, and AI API configurations
User authentication system with JWT tokens, password hashing (Argon2), and API key generation
AI model service integrations with prompt templating and safe JSON response parsing
Safe JSON parsing utility to handle AI response parsing with fallback support

Reviewed changes

Copilot reviewed 57 out of 65 changed files in this pull request and generated 39 comments.

Show a summary per file

File	Description
`types/express.d.ts`	Extends Express Request to include optional user property for authentication
`tsconfig.json`	TypeScript configuration with strict mode and ES2022 target
`src/validator/env.ts`	Environment variable validation schema using Zod
`src/utils/safeJson.util.ts`	Safe JSON parser utility with error handling and fallback
`src/utils/__test__/safeJson.util.test.ts`	Unit tests for safe JSON parsing utility
`src/utils/SendResponse.utils.ts`	Standardized success/error response utility
`src/server.ts`	Main server initialization with MongoDB connection
`src/routes/app.ts`	Central route aggregation and middleware setup
`src/config/mongoose.connection.ts`	MongoDB connection handler with retry logic
`src/constant/env.constant.ts`	Environment constant exports with validation
`src/constant/Status.constant.ts`	HTTP status code enum definitions
`src/api/v1/user/*`	Complete user module (model, service, controller, routes, middleware, tests)
`src/api/v1/AiModelConfig/*`	AI model configuration management system
`src/api/v1/DataSet/v1/*`	Dataset processing with CSV loading and AI extraction
`src/api/v1/Ai-model/Google/*`	Google Gemini AI integration
`src/api/v1/Ai-model/Groq/*`	Groq AI integration
`src/api/v1/Ai-model/Ollama/*`	Ollama local AI integration with embeddings
`src/Template/v1/*`	AI prompt template system with dynamic variable substitution
`src/data/Sample.csv`	Sample dataset for testing data processing
`src/doc/*.json`	Postman API collection for testing endpoints
`package.json`	Project dependencies and scripts
`jest.config.ts`	Jest testing configuration
`.prettierrc`, `.prettierignore`	Code formatting configuration
`.gitignore`	Git ignore patterns
`.env.example`	Environment variable template
`setup-cloudflare.sh`	Cloudflare tunnel setup script
`a.md`	Documentation for training data and tunneling

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-01-03T17:56:58Z

src/api/v1/Ai-model/Groq/Groq.service.ts

+import { ChartWithGroq_type } from './Groq.type'
+import { env } from '../../../../constant/env.constant'
+import AiTemplate from '../../../../Template/v1/Ai.template'
+import { safeParse } from '../../../../utils/safeJson.util'
+
+class GroqService {
+  public async ChartWithGroq(data: ChartWithGroq_type): Promise<any> {


The imported type 'ChartWithGroq_type' should be 'ChatWithGroq_type' to correctly represent a chat interaction.

Suggested change

import { ChartWithGroq_type } from './Groq.type'

import { env } from '../../../../constant/env.constant'

import AiTemplate from '../../../../Template/v1/Ai.template'

import { safeParse } from '../../../../utils/safeJson.util'

class GroqService {

public async ChartWithGroq(data: ChartWithGroq_type): Promise<any> {

import { ChatWithGroq_type } from './Groq.type'

import { env } from '../../../../constant/env.constant'

import AiTemplate from '../../../../Template/v1/Ai.template'

import { safeParse } from '../../../../utils/safeJson.util'

class GroqService {

public async ChartWithGroq(data: ChatWithGroq_type): Promise<any> {

Copilot · 2026-01-03T17:56:58Z

src/api/v1/AiModelConfig/AiModelConfig.controller.ts

+      if (!existingConfig) {
+        CreateConfig = await AiModelConfigService.setupAiModelConfig({
+          userId: String(FindUserByToken._id),
+          agents: [{} as IAgent],


The setupAiModelConfig method creates an initial config with an empty agent object {} as IAgent, which will fail validation. The agent schema requires provider, type, and model fields. This should either not create an initial empty agent or should handle the initial config creation differently.

Suggested change

agents: [{} as IAgent],

agents: [] as IAgent[],

Copilot · 2026-01-03T17:56:59Z

src/api/v1/Ai-model/Ollama/Ollama.controller.ts

+
+      SendResponse.success(res, 'AI response generated successfully', Ai_Response, 200)
+    } catch (err: any) {
+      console.log('err', err)


The console.log statement for error 'err' should be removed or replaced with proper error logging. Production code should use a logging framework instead of console.log for errors.

Copilot · 2026-01-03T17:56:59Z

src/api/v1/Ai-model/Ollama/Ollama.service.ts

+      if (typeof result === 'string') {
+        return safeParse(result, result)


The safeParse usage here doesn't make sense. When the response is already a string, calling safeParse with the string as both input and fallback will either parse it as JSON or return the original string. This means non-JSON string responses will be passed through unchanged, which may not be the intended behavior. Consider clarifying the expected response format.

Copilot · 2026-01-03T17:56:59Z

a.md

@@ -0,0 +1,11 @@
+To Train al model we use Csv , Excel , pdf


The spelling error in 'al model' should be 'AI model' (capital 'AI').

Suggested change

To Train al model we use Csv , Excel , pdf

To Train AI model we use Csv , Excel , pdf

Copilot · 2026-01-03T17:57:06Z

src/api/v1/user/user.service.ts

+  return Promise.resolve(crypto.randomBytes(32).toString('hex'))
+}
+
+const hasedRawKey = (rawKey: string): Promise<string> => {


The function name 'hasedRawKey' contains a spelling error. It should be 'hashedRawKey' (with two 'h's). The function performs hashing, so the correct past tense of 'hash' is 'hashed'.

Copilot · 2026-01-03T17:57:06Z

src/config/mongoose.connection.ts

+    })
+}
+
+export default mongooseConection


The exported default 'mongooseConection' has a spelling error. It should be 'mongooseConnection' to match the corrected function name.

Copilot · 2026-01-03T17:57:07Z

src/api/v1/Ai-model/Groq/Groq.controller.ts

+    try {
+      const { model, apiKey, message } = req.body
+
+      const chat = await GroqService.ChartWithGroq({ model, apiKey, message })


The service method call 'ChartWithGroq' should be 'ChatWithGroq' to correctly represent a chat interaction.

Suggested change

const chat = await GroqService.ChartWithGroq({ model, apiKey, message })

const chat = await GroqService.ChatWithGroq({ model, apiKey, message })

Copilot · 2026-01-03T17:57:07Z

src/api/v1/AiModelConfig/AiModelConfig.service.ts

+  async removeAgent(userId: string, _agentId: string): Promise<IAiModelConfig | null> {
+    const config = (await AiModelConfig.findOne({ userId }).exec()) as any
+
+    if (!config) {
+      throw new Error('AI Model Config not found for the user.')
+    }
+
+    return await config.save()


The removeAgent method doesn't actually remove the agent. After finding the config, it immediately calls save() without removing any agent from the agents array. The implementation is incomplete and should include logic to filter out the agent with the given _agentId before saving.

Copilot · 2026-01-03T17:57:07Z

src/api/v1/DataSet/v1/DataSet.controller.ts

+
+      const Prepare_dataSet = await DataSetService.Prepare_DataSet(documents)
+
+      const parsed = typeof Prepare_dataSet === 'string' ? safeParse(Prepare_dataSet, Prepare_dataSet) : Prepare_dataSet


The safeParse usage is redundant. When Prepare_dataSet is already a string, parsing it with itself as the fallback means if parsing fails, it returns the unparsed string. When it's not a string, it's returned as-is. This logic should be clearer about what format is expected and handle parsing failures appropriately.

morningstarxcdcode · 2026-01-04T06:14:20Z

#42

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

abhishek-nexgen-dev · 2026-01-06T06:51:04Z

@morningstarxcdcode

fix(safe-parse): add safe JSON parse utility and apply to AI services…

75fc9c8

… and dataset; add tests

Copilot AI review requested due to automatic review settings January 3, 2026 17:41

Copilot started reviewing on behalf of morningstarxcdcode January 3, 2026 17:41 View session

gemini-code-assist bot reviewed Jan 3, 2026

View reviewed changes

Copilot AI reviewed Jan 3, 2026

View reviewed changes

morningstarxcdcode mentioned this pull request Jan 3, 2026

fix(safe-parse): add safe JSON parse utility and apply to AI services and dataset processing #43

Closed

morningstarxcdcode mentioned this pull request Jan 4, 2026

🐛 Unhandled JSON parsing in AI services and dataset processing #42

Open

Apply suggestion from @Copilot

5f6b601

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

		if (typeof result === 'string') {
		return safeParse(result, result)

	To Train al model we use Csv , Excel , pdf
	To Train AI model we use Csv , Excel , pdf

	const chat = await GroqService.ChartWithGroq({ model, apiKey, message })
	const chat = await GroqService.ChatWithGroq({ model, apiKey, message })


		const Prepare_dataSet = await DataSetService.Prepare_DataSet(documents)

		const parsed = typeof Prepare_dataSet === 'string' ? safeParse(Prepare_dataSet, Prepare_dataSet) : Prepare_dataSet

fix(safe-parse): add safe JSON parse utility and apply to AI services… #49

Are you sure you want to change the base?

fix(safe-parse): add safe JSON parse utility and apply to AI services… #49

Conversation

morningstarxcdcode commented Jan 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

gemini-code-assist bot commented Jan 3, 2026

Summary of Changes

Highlights

Footnotes

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 3, 2026

Choose a reason for hiding this comment

Uh oh!

morningstarxcdcode commented Jan 4, 2026

Uh oh!

abhishek-nexgen-dev commented Jan 6, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

morningstarxcdcode commented Jan 3, 2026 •

edited

Loading