Qoder IDE is an AI-native integrated development environment that brings code editing, project understanding, and agent collaboration into one workspace. Write code yourself, ask AI to revise a selected section, or delegate a change across files to Agent. Review the diff and run your checks before keeping the result.
It is a standalone IDE: install it and open a project without adding a separate AI coding plugin. See What is Qoder to choose among the other clients.
Choose a collaboration mode based on the task and the amount of code involved:
For example, to add parameter validation to an existing API, first use Ask to locate the entry point, callers, and tests. Then ask Agent to update the implementation and add boundary tests. Use Inline Chat for local adjustments, and finish by checking the diff and test results.
Provide the relevant files, expected behavior, and constraints along with your request. This helps keep changes within the intended scope.
For changes spanning several modules, use Planning to agree on scope, implementation steps, and acceptance criteria before execution. Review tool calls and file changes as work proceeds, and provide further direction when needed.
Combine these capabilities for recurring or specialized tasks:
After Agent finishes, check that its changes meet your requirements before keeping them. Use Agent change review to inspect diffs across files and accept or reject changes. Verify behavior with the project's tests, build, or runtime checks. Confirmation requirements for terminal and tool execution depend on the current policy.
Use the model selector to see the models and options available to your account. Model availability, features, and usage can depend on your account, plan, and organization settings; see billing for details.
Work in the Editor
Choose a collaboration mode based on the task and the amount of code involved:
| Your task | Capability | How to work with it |
|---|---|---|
| Continue writing code or make a related edit | NEXT | Review context-aware suggestions while editing, then press Tab to accept. |
| Explain or revise a small code section | Inline Chat | Select code and describe the change; review and accept or reject the proposed edit in place. |
| Understand a project, locate an implementation, or compare approaches | Ask | Ask a question and attach relevant context. Ask provides explanations and suggestions without directly editing files or running terminal commands. |
| Implement a feature, fix a bug, or add tests | Agent | Describe the goal and acceptance criteria. Agent can search code, change multiple files, and call tools; review its results. |
Give AI your project context and requirements
Provide the relevant files, expected behavior, and constraints along with your request. This helps keep changes within the intended scope.
- Task context: Use @Mention to reference files, folders, or rules. Codebase indexing helps find relevant implementations.
- Team conventions: Put coding standards and test requirements in Rules, using the appropriate scope and activation mode instead of repeating them in every request.
- Project knowledge: Repo Wiki and Knowledge Cards organize architecture, technology, and code specifications. Memory retains preferences and project experience. Check generated knowledge against the current code and update status before relying on it.
Plan complex work and extend execution
For changes spanning several modules, use Planning to agree on scope, implementation steps, and acceptance criteria before execution. Review tool calls and file changes as work proceeds, and provide further direction when needed.
Combine these capabilities for recurring or specialized tasks:
- Skills and custom commands: reuse workflows or prompts for tasks such as reviews and testing.
- Custom agents: configure responsibilities, instructions, and available tools for a particular task.
- MCP: connect external tools and services needed for development.
- Hooks: run custom handlers, such as checks or notifications, when supported execution events occur.
Review changes and verify results
After Agent finishes, check that its changes meet your requirements before keeping them. Use Agent change review to inspect diffs across files and accept or reject changes. Verify behavior with the project's tests, build, or runtime checks. Confirmation requirements for terminal and tool execution depend on the current policy.
Choose models and accounts
Use the model selector to see the models and options available to your account. Model availability, features, and usage can depend on your account, plan, and organization settings; see billing for details.
Get started
- Download the installer for your operating system from the Qoder IDE section.
- Follow the Quick Start to install, sign in, open a local project, and try a code suggestion and an Ask or Agent conversation.
- Check the IDE release notes for new features and version changes.