DevFlow Support

How do I use local Llama models with DevFlow?

To use local Llama models:

• Install Ollama on your machine (https://ollama.ai).
• In DevFlow, go to Settings and enter the Ollama endpoint (usually http://localhost:11434).
• When creating or editing an agent, select an Ollama model (e.g., llama2).

Local processing ensures your code never leaves your machine.

How do I set up repositories for PR reviews?

To set up repositories:

• Go to the Settings page in DevFlow.
• Find the "Repositories" section.
• Enter your repositories in the format "org/repo", one per line.

These repositories will now be available for PR reviews.

How do I add my GitHub token?

To add your GitHub token:

1. Go to the Settings page in DevFlow.
2. Locate the "GitHub Configuration" section.
3. Enter your GitHub personal access token in the "GitHub Token" field.
4. Save your changes.

Ensure your token has the necessary permissions for repository access.

How do I start a PR AI code review?

To start a PR AI code review:

1. Go to the "Review History" view.
2. Tap the "+" button to start a new review.
3. Select the repository and enter the PR number.
4. Choose the AI model for the review.
5. Click "Review PR" to begin the process.

The AI will analyze the PR and provide feedback.

How do I start a general code review?

To start a general code review:

1. Navigate to the "Agents" tab.
2. Select a code review agent.
3. Paste your code into the input field.
4. Click "Run" to start the review process.

The AI will analyze your code and provide feedback.

Can I use my own AI API keys?

Yes, DevFlow supports using your own API keys for services like OpenAI and Claude. You can enter these in the Settings page of the app. This allows you to use your existing AI service accounts.

How do I create a custom AI agent?

To create a custom AI agent:

1. Go to the "Agents" tab.
2. Click the "+" button.
3. Fill in the required information including name, AI model, and template.
4. Save your new agent.

Custom agents allow you to tailor the AI's behavior to your specific needs.

Is my code secure when using DevFlow?

DevFlow prioritizes your code's security:

• When using local models (like Ollama), your code never leaves your machine.
• When using built-in models, processing is done locally.
• If using external AI services (like OpenAI), code is sent securely according to their privacy policies.
• We never store your code or API keys on our servers.