Skip to content

Cld2labs blueprints #2403

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
gopal-raj-suresh wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Cld2labs blueprints #2403

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
f74f0d1
Add Cloud2Labs blueprints
gopal-raj-suresh Jan 21, 2026
b069120
Update repository URLs to opea-project/GenAIExamples for all blueprints
gopal-raj-suresh Jan 21, 2026
7faa48c
Fix directory references in README deployment instructions
gopal-raj-suresh Jan 21, 2026
a4d564e
Merge branch 'opea-project:main' into cld2labs-blueprints
gopal-raj-suresh Jan 21, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
341 changes: 341 additions & 0 deletions CodeGeneration/README.MD
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,341 @@
## Continue VS Code Extension

AI-powered coding assistant integrated with enterprise GenAI Gateway. Provides code completion, chat, and code editing capabilities using locally deployed language models.

## Table of Contents

- [Project Overview](#project-overview)
- [Features](#features)
- [Architecture](#architecture)
- [Prerequisites](#prerequisites)
- [Quick Start](#quick-start)
- [Configuration](#configuration)
- [Usage](#usage)
- [Advanced Features](#advanced-features)
- [Troubleshooting](#troubleshooting)

---

## Project Overview

Continue VS Code Extension enables developers to leverage enterprise-deployed Llama 3.2 3B model for code assistance through GenAI Gateway. Provides autocomplete, chat, and editing capabilities with Keycloak authentication.

---

## Features

**Autocomplete**
- Real-time code completion
- Multiline code generation
- Context-aware suggestions
- Configurable debounce and timeout

**Chat Mode**
- Interactive Q&A about code
- Code explanations
- Problem-solving assistance
- Keyboard shortcut: `Ctrl+L`

**Edit Mode**
- Targeted code modifications
- Inline transformations
- Context-aware refactoring
- Keyboard shortcut: `Ctrl+I`

---

## Architecture

```mermaid
graph TB
subgraph "Developer Workstation"
A[VS Code IDE]
B[Continue Extension]
end

subgraph "Enterprise Infrastructure"
C[GenAI Gateway<br/>LiteLLM]
D[Keycloak<br/>OAuth2 Auth]
end

subgraph "Kubernetes Cluster"
E[AI Model Pods<br/>Llama 3.2 3B]
end

A -->|Code Context| B
B -->|HTTPS Request<br/>API Key| C
C -->|Validate Token| D
D -->|Auth Success| C
C -->|Inference Request| E
E -->|Model Response| C
C -->|Response| B
B -->|Display Results| A

style A fill:#e1f5ff
style B fill:#fff4e1
style C fill:#f0e1ff
style D fill:#ffe1e1
style E fill:#e1ffe1
```

Developer types code in VS Code. Continue extension sends authenticated request to GenAI Gateway. Gateway validates credentials with Keycloak and routes to model. Model generates response. Result displayed in VS Code.

---

## Prerequisites

### System Requirements

- VS Code (latest stable version)
- GenAI Gateway access with Keycloak authentication
- API key from Gateway administrator

### Verify VS Code Installation

```bash
code --version
```

---

## Quick Start

### Install Continue Extension

1. Open VS Code
2. Press `Ctrl+Shift+X` to open Extensions
3. Search for "Continue"
4. Install: **Continue - open-source AI code agent**
5. Publisher: **Continue**

Command line installation:

```bash
code --install-extension Continue.continue
```

### Configure Extension

1. Press `Ctrl+Shift+P`
2. Type "Continue: Open config.yaml"
3. Replace contents with configuration below
4. Update `apiBase` and `apiKey` with your credentials
5. Reload VS Code: `Ctrl+Shift+P` → "Developer: Reload Window"

---

## Configuration

Configuration file location:

**Windows:**
```
C:\Users\<username>\.continue\config.yaml
```

**macOS/Linux:**
```
~/.continue/config.yaml
```

### Basic Configuration

```yaml
name: GenAI Gateway Config
version: 1.0.0
schema: v1

tabAutocompleteOptions:
multilineCompletions: "always"
debounceDelay: 2500
maxPromptTokens: 100
prefixPercentage: 1.0
suffixPercentage: 0.0
maxSuffixPercentage: 0.0
modelTimeout: 5000
showWhateverWeHaveAtXMs: 2000
useCache: true
onlyMyCode: true
useRecentlyEdited: true
useRecentlyOpened: true
useImports: true
transform: true
experimental_includeClipboard: false
experimental_includeRecentlyVisitedRanges: true
experimental_includeRecentlyEditedRanges: true
experimental_includeDiff: true
disableInFiles:
- "*.md"

models:
- name: "Llama 3.2 3B"
provider: openai
model: "meta-llama/Llama-3.2-3B-Instruct"
apiBase: "https://api.example.com/v1"
apiKey: "your-api-key-here"
ignoreSSL: true
contextLength: 8192
completionOptions:
maxTokens: 1024
temperature: 0.3
stop:
- "\n\n"
- "def "
- "class "
requestOptions:
maxTokens: 1024
temperature: 0.3
autocompleteOptions:
maxTokens: 256
temperature: 0.2
stop:
- "\n\n\n"
- "# "
roles:
- chat
- edit
- apply
- autocomplete
promptTemplates:
autocomplete: "{{{prefix}}}"

useLegacyCompletionsEndpoint: true
experimental:
inlineEditing: true
allowAnonymousTelemetry: false
```

### Required Updates

1. **apiBase**: Your GenAI Gateway URL with `/v1` suffix
2. **apiKey**: API key from Gateway administrator
3. **model**: Exact model name `meta-llama/Llama-3.2-3B-Instruct`

For detailed configuration options and advanced setup, refer to [SETUP_GUIDE.md](./SETUP_GUIDE.md).

### Verify Configuration

```bash
export API_KEY="your-api-key-here"
export API_BASE="https://api.example.com/v1"
```

```bash
curl -k $API_BASE/models \
-H "Authorization: Bearer $API_KEY"
```

```bash
curl -k $API_BASE/chat/completions \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "meta-llama/Llama-3.2-3B-Instruct",
"messages": [{"role": "user", "content": "What is Python?"}],
"max_tokens": 50
}'
```

---

## Usage

### Agent mode

**How to Use:**
1. Open Continue sidebar
2. Switch to Agent mode
3. Give task instruction
4. Review and approve file operations
5. Verify results

**Preview:** Requested "Create a FastAPI application with two routes". The model generated the code and created a new file with the complete implementation including imports, app initialization, and route definitions.

![Agent Mode Demo](./src/agent-mode.png)

![Agent Mode Demo](./src/agent-mode.gif)

### Autocomplete

**How to Use:**
1. Start typing code
2. Pause 3 seconds
3. Accept with `Tab` or reject by continuing to type

Enable/disable via "Continue" button in status bar.

**Preview:** Started typing to create an endpoint for the sample FastAPI application and paused. The model generated the code for the endpoint and provided a prompt to accept or reject the code.

![Autocomplete Demo](./src/autocomplete-demo.png)

![Autocomplete Demo](./src/autocomplete-demo.gif)



---

### Chat Mode

**How to Use:**
1. Press `Ctrl+L`
2. Type question
3. Press Enter
4. Review response

Context providers:
- Highlight code for automatic inclusion
- `@Files` - Reference specific files
- `@Terminal` - Include terminal output

**Preview:** Asked how does FastAPI handles request validation in the current file and received the response with a suggestion, which can be viewed in the screenshots below

![Chat Mode Demo](./src/chat-mode.png)

![Chat Mode Demo](./src/chat-mode-1.png)


![Chat Mode Demo](./src/chat-mode.gif)


---

### Edit Mode

**How to Use:**
1. Highlight code
2. Press `Ctrl+I`
3. Type instruction
4. Review diff
5. Accept or reject


**Preview:** Highlighted the code file and provided a prompt to "convert every endpoint to async". The model generated a difference showing the original code and proposed changes and provided a prompt to accept or reject the code.

![Edit Mode Demo](./src/edit-mode-demo.png)



![Edit Mode Demo](./src/edit-mode.gif)


---

## Advanced Features

**Custom Rules**
- Define custom system prompts and context for specific project needs
- Control AI behavior with project-specific guidelines

**MCP Servers**
- Extend functionality with Model Context Protocol servers
- Add custom tools and external integrations

For detailed setup instructions on creating custom rules and MCP servers, refer to [SETUP_GUIDE.md](./SETUP_GUIDE.md).

---

## Troubleshooting

For comprehensive troubleshooting guidance, common issues, and solutions, refer to:

[Troubleshooting Guide - TROUBLESHOOTING.md](./TROUBLESHOOTING.md)
Loading
Loading