First Run

After installing NexusCore CLI or NexusIDE, start your first AI agent session.

Prerequisites

You need at least one model provider configured. If you haven't set one up yet, see Connecting a Model.

TIP

NexusIDE includes a first-run setup wizard that walks you through model configuration automatically. Just launch the IDE and follow the prompts.

NexusCore CLI

Interactive Chat

Start a conversation with the agent:

nexus-cli chat

The agent enters an interactive session where you can type messages and receive responses. Type exit or press Ctrl+C to end the session.

One-Shot Tasks

Run a single task and get the result:

nexus-cli run "Create a Python hello world script"

The agent executes the task, uses tools as needed (file creation, terminal commands), and prints the result.

Explore Available Tools

See what tools the agent can use:

nexus-cli tools list

Example output:

Available tools (22):
  file_read        Read file contents
  file_write       Create or overwrite a file
  file_list        List directory contents
  file_delete      Delete a file
  text_search      Search for text patterns
  terminal_run     Execute a shell command
  web_search       Search the web
  web_fetch        Fetch a URL
  code_analyze     Analyze code structure
  git_status       Show git status
  git_commit       Create a git commit
  git_diff         Show git diff
  ...

NexusIDE

Open the Agent Panel

1. Launch NexusIDE and open a project folder
2. Click the NexusCore icon in the activity bar (left sidebar)
3. The Agent Panel opens — type a message to start chatting

Agent Modes

The chat panel includes a mode selector with three options:

Mode	Description	Use Case
Ask	Single-turn Q&A, no file edits	Quick questions, explanations
Edit	Agent can read and edit files with your approval	Focused code changes
Agent	Autonomous multi-step execution with tool use	Complex tasks, refactoring

Inline Edits

For quick, targeted changes:

1. Select code in the editor
2. Press Ctrl+I / Cmd+I
3. Describe the change
4. Review the inline diff and accept or reject

Status Bar

The bottom status bar shows your current state:

• NexusCore · Free — Not signed in, Core tier features available
• NexusCore · Pro — Signed in with Pro tier
• NexusCore · Studio — Signed in with Studio tier
• The active model name appears next to the status icon (e.g., qwen2.5-coder-32b)

Core Features (Free)

The Core tier gives you full access to everything you need for local AI coding:

• 22 built-in tools — File operations, web search, code analysis, git, terminal, and more
• Swarm mode — Decompose complex tasks across parallel sub-agents
• Skills — Extend the agent with reusable automation patterns
• MCP support — Connect to Model Context Protocol servers for external tools
• Plugins — Add custom tool integrations
• Checkpoints — Save and restore workspace state before and after agent changes
• Memory — Persistent context across sessions so the agent remembers your preferences
• BYOM — Bring your own model from any provider — local or cloud

No sign-in required. Core features work fully offline.

What's Next

• Connecting a Model — Set up local or cloud LLM providers
• Authentication — Sign in to unlock Pro and Studio features
• NexusIDE Guide — Deep dive into IDE features
• Skills Guide — Create and install agent skills