---
title: "Tips & Troubleshooting"
id: "354"
type: "page"
slug: "tips"
published_at: "2026-05-19T11:30:47+00:00"
modified_at: "2026-06-11T14:20:24+00:00"
url: "https://xedant.com/code/docs/tips"
markdown_url: "https://xedant.com/code/docs/tips.md"
excerpt: "Practical tips for getting the most out of Xedant Code, answers to common questions, and…"
---

# Tips & Troubleshooting

[https://xedant.com/code/docs/tips.md](https://xedant.com/code/docs/tips.md)

Practical tips for getting the most out of Xedant Code, answers to common questions, and troubleshooting steps for when things don’t work as expected.

## Tips for Effective AI Coding

- **Be specific** — The more detail you give, the better the AI understands your request. Instead of “fix the button”, say “the submit button on the contact form doesn’t validate email format — add email validation using a regex pattern”
- **Break large tasks into smaller ones** — Instead of “build the whole app”, start with one feature at a time. The AI works best with focused, well-defined requests
- **Use skills** — Pick the right skill for the task using the skills selector. A “code review” skill will give different results than a “development” skill for the same code
- **Iterate** — If the first response isn’t perfect, ask for refinements rather than starting over. “Make the heading larger and change the color to blue” is faster than re-describing the whole requirement
- **Provide context** — Mention relevant file names, function names, or patterns you want the AI to follow. “Update the formatDate function in src/utils/date.ts to handle European date format”

## Common Questions

**How do I connect my own project?**

Reconfigure your [environment variables](/code/docs/environments)
 (project path, git URL, credentials) and restart Xedant Code. See the [Getting Started](/code/docs/getting-started)
 section and the installation guides for [Windows](/code/windows)
 or [Docker](/code/docker)
 for details.

**Can I use multiple AI models?**

Yes. Create multiple [environments](/code/docs/environments)
 in Settings, each with a different model and API key. Switch between them using the environment selector in the chat input.

**What happens when the AI makes a mistake?**

The AI sees build errors and can fix them automatically if [AutoFix](/code/docs/autofix)
 is enabled. You can also tell the AI what went wrong and ask it to correct the issue. The AI reads error messages from the [Build panel](/code/docs/build-deploy)
 and uses them as context for fixes.

**Can I share skills between projects?**

Yes. Skills are stored in a separate git repository. Point multiple Xedant Code instances to the same skills repo URL to share skills across projects and team members. See the [Skills](/code/docs/skills)
 section for details.

**How do I back up my data?**

Your project files are in your git repository — commit and push regularly. Chat history and settings are stored in the local database. Back up the database file or configure a persistent database for production use.

**Is my data private?**

Your code runs locally in your own container or instance. Chat messages are sent to the AI provider you configured (such as Anthropic or OpenAI). Analytics data stays local unless you configure ClickHouse push for centralized reporting.

## Connection Issues

**Can’t reach the AI:** Check that your API key is valid and that you have available API credits with your provider. Verify the API key in your [environment configuration](/code/docs/environments)
.

**Can’t push or pull Git changes:** Verify your git credentials (login, personal access token) are configured in your environment variables. See the [Git](/code/docs/git)
 section for details.

**Container not starting:** Check Docker logs for startup errors. Common causes include incorrect [environment variables](/code/docs/environments)
, port conflicts, or missing configuration files.

**Port already in use:** Another service is using the port Xedant Code is configured to listen on. Change the port in your configuration or stop the conflicting service.

## Performance Tips

- **Large projects** — Xedant Code handles large codebases, but performance depends on your container’s allocated resources. Increase memory and CPU if you’re working with very large projects
- **Long conversations** — Very long chat sessions use more tokens and may slow down. Start a new session when the conversation context is no longer relevant to your current task
- **Build speed** — If you don’t need the full development environment, use the minimal Docker image for faster builds and deploys
- **Database** — For better performance with many chat sessions over time, consider using PostgreSQL instead of SQLite

## Getting Help

If you’ve tried the troubleshooting steps above and still need assistance:

- **Telegram** — Contact `@maxedant` for direct support (fastest response)
- **Email** — Send a message to `support@xedant.com`
- **FAQ page** — Visit the [FAQ](/code/faq) for additional answers to common questions

Response time is typically within 12 hours, including weekends. When contacting support, include a description of the issue, any error messages, and steps to reproduce the problem.

**[← Shortcuts](/code/docs/shortcuts)**

**[Tips & Troubleshooting →](/code/docs/tips)**