Every Code AI Agent

Every Code (Code for short) is a fast, local coding agent for your terminal. It's a community-driven fork focused on real developer ergonomics: browser integration, multi-agents, theming, and reasoning control.

What This Skill Does

This skill helps you install, configure, and use Every Code CLI to orchestrate AI-powered coding workflows with multi-agent support, automated code reviews, and browser integration.

Key Features

Installation Steps

1. **Install Every Code globally**

```bash

npm install -g @just-every/code

```

Note: If another tool provides a `code` command (e.g. VS Code), use `coder` instead.

2. **Authenticate** (choose one method)

**Option A: Sign in with ChatGPT** (Plus/Pro/Team)

```bash

code

Select "Sign in with ChatGPT" when prompted

```

**Option B: API key** (usage-based)

```bash

export OPENAI_API_KEY=your_key_here

code

```

3. **Install companion CLIs (optional but recommended)**

```bash

Ensure Node.js 20+ is available

npm install -g n

export N_PREFIX="$HOME/.n"

export PATH="$N_PREFIX/bin:$PATH"

n 20.18.1

Install companion CLIs

export npm_config_prefix="${npm_config_prefix:-$HOME/.npm-global}"

mkdir -p "$npm_config_prefix/bin"

export PATH="$npm_config_prefix/bin:$PATH"

npm install -g @anthropic-ai/claude-code @google/gemini-cli @qwen-code/qwen-code

Verify installations

claude --version

gemini --version

qwen --version

```

Add these exports to your shell profile (.bashrc, .zshrc) for persistence:

```bash

export N_PREFIX="$HOME/.n"

export PATH="$N_PREFIX/bin:$PATH"

export npm_config_prefix="${npm_config_prefix:-$HOME/.npm-global}"

export PATH="$npm_config_prefix/bin:$PATH"

```

Essential Commands

Auto Drive Orchestration

```bash

Hand off multi-step tasks

code /auto "Refactor the auth flow and add device login"

Check status

code /auto status

```

Multi-Agent Commands

```bash

Plan changes (Claude, Gemini, GPT-5 consensus)

code /plan "Stop the AI from ordering pizza at 3AM"

Solve problems (fastest agent wins)

code /solve "Why does deleting one user drop the whole database?"

Write code (consensus approach)

code /code "Show dark mode when I feel cranky"

```

Browser Integration

```bash

Connect to external Chrome with CDP

code /chrome # Auto-detect port

code /chrome 9222 # Specific port

Use internal headless browser

code /browser

code /browser https://example.com

```

Configuration Commands

```bash

Switch themes

code /themes

Adjust reasoning level

code /reasoning low|medium|high

Change models

code /model

Start fresh conversation

code /new

```

CLI Options

```bash

code [options] [prompt]

Options:

--model <name> Override model (e.g. gpt-5.1)

--read-only Prevent file modifications

--no-approval Skip approval prompts (caution!)

--config <key=val> Override config values

--oss Use local open source models

--sandbox <mode> Set sandbox level

--debug Log API requests/responses

--help Show help

--version Show version

```

Project Memory Setup

Create `AGENTS.md` or `CLAUDE.md` in your project root:

```markdown

Project Context

This is a React TypeScript application with:

Key files:

```

Configuration

Main config: `~/.code/config.toml`

```toml

Model settings

model = "gpt-5.1"

model_provider = "openai"

Behavior

approval_policy = "on-request" # untrusted | on-failure | on-request | never

model_reasoning_effort = "medium" # low | medium | high

sandbox_mode = "workspace-write"

Theme

[tui.theme]

name = "light-photon"

Model profiles

[profiles.gpt-5]

model = "gpt-5.1"

model_provider = "openai"

approval_policy = "never"

model_reasoning_effort = "high"

```

Environment Variables

MCP Configuration

Extend capabilities with Model Context Protocol. Add to `~/.code/config.toml`:

```toml

[mcp_servers.filesystem]

command = "npx"

args = ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/project"]

[mcp_servers.database]

command = "npx"

args = ["-y", "@modelcontextprotocol/server-postgres", "postgresql://localhost/mydb"]

```

CI/Automation Mode

```bash

Automated task execution

code --no-approval "run tests and fix any failures"

Read-only analysis

code --read-only "analyze code quality and generate report"

JSON output for parsing

code --config output_format=json "list all TODO comments"

```

Usage Examples

**Example 1: Auto Drive multi-step task**

```bash

code /auto "Add user profile page with avatar upload and bio editing"

```

**Example 2: Multi-agent problem solving**

```bash

code /solve "Memory leak in the WebSocket connection handler"

```

**Example 3: Browser-integrated debugging**

```bash

code /chrome 9222

Then in chat: "Navigate to localhost:3000 and debug the login form validation"

```

**Example 4: Code review workflow**

```bash

Auto Review runs in background on code changes

code /auto "Implement payment processing with Stripe"

Review runs automatically when code changes

```

Important Notes

Constraints

Quick Start Template

```bash

1. Install

npm install -g @just-every/code

2. Set API key or sign in

export OPENAI_API_KEY=your_key

3. Create project context

cat > AGENTS.md << 'EOF'

My Project

[Describe your project architecture and key files]

EOF

4. Run your first auto task

code /auto "Add health check endpoint to the API"

```

Resources