Quickstart

Sign in, send your first prompt to start a Task and its first agent, and review the result.

This guide gets you from a fresh Traycer install to one useful coding-agent interaction. If you have not installed Traycer yet, start with Install.

This is the Desktop path. If you are using the IDE extension, use the IDE Extension docs.

Before You Start

You need:

• Traycer installed.
• A coding agent you can use, such as Claude Code or Codex with your own account, or the Traycer provider, which uses Traycer credits. See Agents & Models.
• Optionally, a local code folder for Traycer to work in.

First Run

1. Sign in

   Open Traycer and click Sign in on the Welcome to Traycer screen. You approve the sign-in in your browser. See Sign In.

2. Finish onboarding

   A short onboarding lets you choose your coding agents and import your existing Claude Code, Codex, and OpenCode chats. See First run and importing.

3. Pick a folder (optional)

   On the start page, click Add folder to choose the project Traycer should work in, then choose where the agent runs: Local, a New worktree, or an Existing worktree. See Worktrees. You can also start without a folder.

4. Send your first prompt

   New conversations start in Full access: the agent runs commands and edits files without asking. To be asked first, pick another mode in the composer’s permission picker before you send, or change the default in Settings › Permissions › Modes › New conversations start in. See Permissions.

   Pick a coding agent, model and permission mode, describe a small, concrete change or investigation, and press Enter. Sending the prompt creates the Task and its first agent together, and the agent starts working.

5. Review the result

   Use the File Tree and Git Diff panels inside the Task to inspect what changed. To roll back a turn’s changes, see Undo a turn.

What You Just Used

Thing	What it means
Workspace folder	The local code folder Traycer is working against.
Task	The top-level container for this work.
Agent	A coding-agent session inside the Task.
Artifact	A durable spec, ticket, story, or review inside a Task.

Next Steps

Learn the Concepts

Understand Tasks, panels, and artifacts.

Work with agents

Understand the Chat and Terminal interfaces, permissions, model selection, attachments, voice input, and per-turn settings.

Set up Agents & Models

Compare the supported coding agents and set up their accounts.

Review changes

Inspect changed files and review diffs before shipping.

Artifact workflow

How a Task holds artifacts, and which planning skills to run for a large feature.