### Recommended First Run: Quickstart Constructor

Source: https://github.com/just-every/code/blob/main/sdk/python/examples/README.md

Run the recommended first examples for a quickstart constructor sanity check, covering both synchronous and asynchronous flows.

```bash
python examples/01_quickstart_constructor/sync.py
```

```bash
python examples/01_quickstart_constructor/async.py
```

--------------------------------

### Install SDK Dependencies

Source: https://github.com/just-every/code/blob/main/sdk/python/examples/README.md

Install SDK dependencies using uv and activate the virtual environment. This is a recommended setup step before running examples.

```bash
uv sync
source .venv/bin/activate
```

--------------------------------

### Clone and Setup Project

Source: https://github.com/just-every/code/blob/main/README.md

Clone the repository, navigate into the directory, and install dependencies.

```bash
git clone https://github.com/just-every/code.git
cd code
npm install
```

--------------------------------

### windowsSandbox/setupStart

Source: https://github.com/just-every/code/blob/main/codex-rs/app-server/README.md

Starts Windows sandbox setup for the selected mode ('elevated' or 'unelevated'). Accepts an optional absolute `cwd` to target setup for a specific workspace. Returns `{ started: true }` immediately and later emits `windowsSandbox/setupCompleted`.

```APIDOC
## windowsSandbox/setupStart

### Description
Starts Windows sandbox setup for the selected mode.

### Method
POST

### Endpoint
`/windowsSandbox/setupStart`

### Parameters
#### Request Body
- **mode** (string) - Required - The mode for sandbox setup (`elevated` or `unelevated`).
- **cwd** (string) - Optional - The absolute path to the workspace directory.

### Response
#### Success Response (200)
- **started** (boolean) - Indicates if the setup process has started.
```

--------------------------------

### Quickstart: Basic SDK Usage

Source: https://github.com/just-every/code/blob/main/sdk/python/README.md

Initialize the Codex client, start a thread, run a prompt, and print the final response and item count. Ensure to use a context manager for proper shutdown.

```python
from codex_app_server import Codex

with Codex() as codex:
    thread = codex.thread_start(model="gpt-5")
    result = thread.run("Say hello in one sentence.")
    print(result.final_response)
    print(len(result.items))
```

--------------------------------

### Example Usage

Source: https://github.com/just-every/code/blob/main/sdk/python/docs/api-reference.md

A basic example demonstrating how to use the Codex SDK to start a thread and get a response.

```APIDOC
## Example

```python
from codex_app_server import Codex

with Codex() as codex:
    thread = codex.thread_start(model="gpt-5.4", config={"model_reasoning_effort": "high"})
    result = thread.run("Say hello in one sentence.")
    print(result.final_response)
```

**Description**: This example shows the typical workflow of initializing the `Codex` client, starting a new thread with specific model configuration, running a task on that thread, and printing the final response.
```

--------------------------------

### Start Realtime Communication with WebRTC - Server Responses

Source: https://github.com/just-every/code/blob/main/codex-rs/app-server/README.md

Example server responses for starting a WebRTC realtime session, including an empty result and an SDP answer for setting the remote description.

```json
{ "id": 40, "result": {} }
```

```json
{ "method": "thread/realtime/sdp", "params": {
    "threadId": "thr_123",
    "sdp": "v=0\r\no=..."
} }
```

--------------------------------

### Run Example Scripts

Source: https://github.com/just-every/code/blob/main/sdk/python/README.md

Execute example Python scripts for the SDK from the command line. These examples demonstrate various functionalities.

```bash
cd sdk/python
python examples/01_quickstart_constructor/sync.py
python examples/01_quickstart_constructor/async.py
```

--------------------------------

### Initialize Skill (Example Usage)

Source: https://github.com/just-every/code/blob/main/code-rs/core/src/skills/assets/samples/skill-creator/SKILL.md

Example command to initialize a skill named 'my-skill' within the 'skills/public' directory.

```bash
scripts/init_skill.py my-skill --path skills/public
```

--------------------------------

### Package Skill (Example Usage)

Source: https://github.com/just-every/code/blob/main/code-rs/core/src/skills/assets/samples/skill-creator/SKILL.md

Example command to package a skill located at 'skills/public/my-skill'.

```bash
python scripts/package_skill.py skills/public/my-skill
```

--------------------------------

### Bootstrap SDK Imports and Runtime Setup

Source: https://github.com/just-every/code/blob/main/sdk/python/notebooks/sdk_walkthrough.ipynb

Initializes the SDK by setting up Python path, ensuring the runtime package is installed, and handling fresh imports. This cell is crucial for preparing the environment before using the SDK.

```python
# Cell 1: bootstrap local SDK imports + pinned runtime package
import os
import sys
from pathlib import Path

if sys.version_info < (3, 10):
    raise RuntimeError(
        f'Notebook requires Python 3.10+; current interpreter is {sys.version.split()[0]}.
')

try:
    _ = os.getcwd()
except FileNotFoundError:
    os.chdir(str(Path.home()))


def _is_sdk_python_dir(path: Path) -> bool:
    return (path / 'pyproject.toml').exists() and (path / 'src' / 'codex_app_server').exists()


def _iter_home_fallback_candidates(home: Path):
    # bounded depth scan under home to support launching notebooks from unrelated cwd values
    patterns = ('sdk/python', '*/sdk/python', '*/*/sdk/python', '*/*/*/sdk/python')
    for pattern in patterns:
        yield from home.glob(pattern)


def _find_sdk_python_dir(start: Path) -> Path | None:
    checked = set()

    def _consider(candidate: Path) -> Path | None:
        resolved = candidate.resolve()
        if resolved in checked:
            return None
        checked.add(resolved)
        if _is_sdk_python_dir(resolved):
            return resolved
        return None

    for candidate in [start, *start.parents]:
        found = _consider(candidate)
        if found is not None:
            return found

    for candidate in [start / 'sdk' / 'python', *(parent / 'sdk' / 'python' for parent in start.parents)]:
        found = _consider(candidate)
        if found is not None:
            return found

    env_dir = os.environ.get('CODEX_PYTHON_SDK_DIR')
    if env_dir:
        found = _consider(Path(env_dir).expanduser())
        if found is not None:
            return found

    for entry in sys.path:
        if not entry:
            continue
        entry_path = Path(entry).expanduser()
        for candidate in (entry_path, entry_path / 'sdk' / 'python'):
            found = _consider(candidate)
            if found is not None:
                return found

    home = Path.home()
    for candidate in _iter_home_fallback_candidates(home):
        found = _consider(candidate)
        if found is not None:
            return found

    return None


repo_python_dir = _find_sdk_python_dir(Path.cwd())
if repo_python_dir is None:
    raise RuntimeError('Could not locate sdk/python. Set CODEX_PYTHON_SDK_DIR to your sdk/python path.')

repo_python_str = str(repo_python_dir)
if repo_python_str not in sys.path:
    sys.path.insert(0, repo_python_str)

from _runtime_setup import ensure_runtime_package_installed

runtime_version = ensure_runtime_package_installed(
    sys.executable,
    repo_python_dir,
)

src_dir = repo_python_dir / 'src'
examples_dir = repo_python_dir / 'examples'
src_str = str(src_dir)
examples_str = str(examples_dir)
if src_str not in sys.path:
    sys.path.insert(0, src_str)
if examples_str not in sys.path:
    sys.path.insert(0, examples_str)

# Force fresh imports after SDK upgrades in the same notebook kernel.
for module_name in list(sys.modules):
    if module_name == 'codex_app_server' or module_name.startswith('codex_app_server.'):
        sys.modules.pop(module_name, None)

print('Kernel:', sys.executable)
print('SDK source:', src_dir)
print('Runtime package:', runtime_version)

```

--------------------------------

### Full openai.yaml Example

Source: https://github.com/just-every/code/blob/main/codex-rs/skills/src/assets/samples/skill-creator/references/openai_yaml.md

A complete example of the openai.yaml configuration file, demonstrating all available fields for defining a skill's interface, dependencies, and policies.

```yaml
interface:
  display_name: "Optional user-facing name"
  short_description: "Optional user-facing description"
  icon_small: "./assets/small-400px.png"
  icon_large: "./assets/large-logo.svg"
  brand_color: "#3B82F6"
  default_prompt: "Optional surrounding prompt to use the skill with"

dependencies:
  tools:
    - type: "mcp"
      value: "github"
      description: "GitHub MCP server"
      transport: "streamable_http"
      url: "https://api.githubcopilot.com/mcp/"

policy:
  allow_implicit_invocation: true
```

--------------------------------

### Run a New Shell in a Basic Sandbox

Source: https://github.com/just-every/code/blob/main/codex-rs/vendor/bubblewrap/README.md

A trimmed-down example demonstrating how to run a new bash shell within a sandbox. This setup reuses the host's /usr directory and configures essential virtual filesystems like /proc and /dev.

```sh
bwrap \
    --ro-bind /usr /usr \
    --symlink usr/lib64 /lib64 \
    --proc /proc \
    --dev /dev \
    --unshare-pid \
    --new-session \
    bash
```

--------------------------------

### Example Skill Frontmatter and Content

Source: https://github.com/just-every/code/blob/main/code-rs/core/src/skills/assets/samples/skill-creator/SKILL.md

An example of a SKILL.md file, including frontmatter with name and description, and basic markdown content for a skill.

```markdown
---
name: pdf-processing
description: Extract text and tables from PDFs; use when PDFs, forms, or document extraction are mentioned.
---

# PDF Processing
- Use pdfplumber to extract text.
- For form filling, see FORMS.md.
```

--------------------------------

### Install Skill from GitHub URL

Source: https://github.com/just-every/code/blob/main/codex-rs/skills/src/assets/samples/skill-installer/SKILL.md

Install a skill by providing its direct GitHub URL. This method allows for installation from a specific branch or commit by including the ref in the URL.

```bash
scripts/install-skill-from-github.py --url https://github.com/<owner>/<repo>/tree/<ref>/<path>
```

--------------------------------

### Run Synchronous Example

Source: https://github.com/just-every/code/blob/main/sdk/python/examples/README.md

Execute a synchronous example script. Ensure you are in the 'sdk/python' directory. The SDK source is used for local imports.

```bash
python examples/<example-folder>/sync.py
```

--------------------------------

### Install Codex App Server Python SDK

Source: https://github.com/just-every/code/blob/main/sdk/python/README.md

Navigate to the SDK directory, install dependencies using uv, and activate the virtual environment.

```bash
cd sdk/python
uv sync
source .venv/bin/activate
```

--------------------------------

### Run Asynchronous Example

Source: https://github.com/just-every/code/blob/main/sdk/python/examples/README.md

Execute an asynchronous example script. Ensure you are in the 'sdk/python' directory. The SDK source is used for local imports.

```bash
python examples/<example-folder>/async.py
```

--------------------------------

### Install Skill from GitHub Repository

Source: https://github.com/just-every/code/blob/main/codex-rs/skills/src/assets/samples/skill-installer/SKILL.md

Install one or more skills from a specified GitHub repository and path. This command supports installing skills from any public or private GitHub repository.

```bash
scripts/install-skill-from-github.py --repo <owner>/<repo> --path <path/to/skill> [<path/to/skill> ...]
```

--------------------------------

### Install Non-Default Local Marketplace

Source: https://github.com/just-every/code/blob/main/codex-rs/skills/src/assets/samples/plugin-creator/references/installing-and-updating.md

Command to install a non-default local marketplace before adding plugins from it.

```bash
codex plugin marketplace add <path-to-marketplace-root>
```

--------------------------------

### Install OpenAI MCP Documentation

Source: https://github.com/just-every/code/blob/main/codex-rs/skills/src/assets/samples/openai-docs/SKILL.md

Run this command to install the OpenAI documentation for the MCP server. If it fails due to permissions, retry with escalated permissions.

```bash
codex mcp add openaiDeveloperDocs --url https://developers.openai.com/mcp
```

--------------------------------

### Install Experimental Skill from GitHub

Source: https://github.com/just-every/code/blob/main/codex-rs/skills/src/assets/samples/skill-installer/SKILL.md

Install an experimental skill from the openai/skills repository. This command targets skills located within the `.experimental` path of the repository.

```bash
scripts/install-skill-from-github.py --repo openai/skills --path skills/.experimental/<skill-name>
```

--------------------------------

### YAML Frontmatter Example for SKILL.md

Source: https://github.com/just-every/code/blob/main/code-rs/core/src/skills/assets/samples/skill-creator/SKILL.md

Provides an example of the required YAML frontmatter for a SKILL.md file, including 'name' and 'description' fields.

```yaml
---
name: my-skill
description: Use when the user requests XYZ or mentions ABC workflows.
---
```

--------------------------------

### View Available Options

Source: https://github.com/just-every/code/blob/main/code-rs/responses-api-proxy/npm/README.md

Run the installed command-line proxy with the --help flag to see all available configuration options and commands.

```bash
node ./bin/codex-responses-api-proxy.js --help
```

--------------------------------

### Async Operations

Source: https://github.com/just-every/code/blob/main/sdk/python/docs/getting-started.md

Utilize the asynchronous client for non-blocking operations. This example shows how to start a thread and run a turn asynchronously.

```python
import asyncio
from codex_app_server import AsyncCodex


async def main() -> None:
    async with AsyncCodex() as codex:
        thread = await codex.thread_start(model="gpt-5.4", config={"model_reasoning_effort": "high"})
        result = await thread.run("Continue where we left off.")
        print(result.final_response)


asyncio.run(main())
```

--------------------------------

### Get Thread Goal

Source: https://github.com/just-every/code/blob/main/codex-rs/app-server/README.md

Use `thread/goal/get` to retrieve the current goal of a thread without modifying it. The example shows a case where the goal is null.

```json
{ "method": "thread/goal/get", "id": 29, "params": { "threadId": "thr_123" } }
{ "id": 29, "result": { "goal": null } }
```

--------------------------------

### Build Project from Source

Source: https://github.com/just-every/code/blob/main/docs/install.md

Clone the repository, install Rust, and build the project. This is the same check performed by CI.

```bash
# Clone the repository and navigate to the workspace root.
git clone https://github.com/just-every/code.git
cd code

# Install the Rust toolchain, if necessary.
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
source "$HOME/.cargo/env"

# Build everything (CLI, TUI, MCP servers). This is the same check CI runs.
./build-fast.sh

# Launch the TUI with a sample prompt.
./target/debug/code -- "explain this codebase to me"
```

--------------------------------

### Run First Turn (Sync)

Source: https://github.com/just-every/code/blob/main/sdk/python/docs/getting-started.md

Initialize the Codex client and run a single turn. This example demonstrates starting a thread, sending a message, and retrieving the final response.

```python
from codex_app_server import Codex

with Codex() as codex:
    server = codex.metadata.serverInfo
    print("Server:", None if server is None else server.name, None if server is None else server.version)

    thread = codex.thread_start(model="gpt-5.4", config={"model_reasoning_effort": "high"})
    result = thread.run("Say hello in one sentence.")

    print("Thread:", thread.id)
    print("Text:", result.final_response)
    print("Items:", len(result.items))
```

--------------------------------

### Run Every Code CLI

Source: https://github.com/just-every/code/blob/main/README.md

Execute the Every Code command-line interface using npx. This is a quick way to start using the tool without a global installation.

```bash
npx -y @just-every/code
```

--------------------------------

### Main Configuration File Example

Source: https://github.com/just-every/code/blob/main/README.md

Example of the main Every Code configuration file (`~/.code/config.toml`). Includes model settings, behavior policies, and theme preferences.

```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"

# UI preferences see THEME_CONFIG.md
[tui.theme]
name = "light-photon"

# Add config for specific models
[profiles.gpt-5]
model = "gpt-5.1"
model_provider = "openai"
approval_policy = "never"
model_reasoning_effort = "high"
model_reasoning_summary = "detailed"

```

--------------------------------

### Control Working Directory and Git Check

Source: https://github.com/just-every/code/blob/main/sdk/typescript/README.md

Configure the working directory for Codex or skip the Git repository check when starting a new thread. This provides flexibility in project setup.

```typescript
const thread = codex.startThread({
  workingDirectory: "/path/to/project",
  skipGitRepoCheck: true,
});
```

--------------------------------

### Initialize New Skill with init_skill.py

Source: https://github.com/just-every/code/blob/main/codex-rs/skills/src/assets/samples/skill-creator/SKILL.md

Use this script to create a new skill directory with a template structure. Specify the skill name, output path, and optionally include resource directories or example files.

```bash
scripts/init_skill.py <skill-name> --path <output-directory> [--resources scripts,references,assets] [--examples]
```

```bash
scripts/init_skill.py my-skill --path "${CODEX_HOME:-$HOME/.codex}/skills"
```

```bash
scripts/init_skill.py my-skill --path "${CODEX_HOME:-$HOME/.codex}/skills" --resources scripts,references
```

```bash
scripts/init_skill.py my-skill --path ~/work/skills --resources scripts --examples
```

--------------------------------

### Basic Thread Interaction Example

Source: https://github.com/just-every/code/blob/main/sdk/python/docs/api-reference.md

Demonstrates starting a new thread with a specific model and configuration, sending a message, and printing the final response. Uses a context manager for the Codex instance.

```python
from codex_app_server import Codex

with Codex() as codex:
    thread = codex.thread_start(model="gpt-5.4", config={"model_reasoning_effort": "high"})
    result = thread.run("Say hello in one sentence.")
    print(result.final_response)
```

--------------------------------

### Initialize and Run Network Proxy Library

Source: https://github.com/just-every/code/blob/main/codex-rs/network-proxy/README.md

Demonstrates how to build and run the NetworkProxy library. The policy decider can be customized to automatically allow commands starting with 'curl '.

```rust
use codex_network_proxy::{NetworkProxy, NetworkDecision, NetworkPolicyRequest};

let proxy = NetworkProxy::builder()
    .http_addr("127.0.0.1:8080".parse()?)
    .policy_decider(|request: NetworkPolicyRequest| async move {
        // Example: auto-allow when exec policy already approved a command prefix.
        if let Some(command) = request.command.as_deref() {
            if command.starts_with("curl ") {
                return NetworkDecision::Allow;
            }
        }
        NetworkDecision::Deny {
            reason: "policy_denied".to_string(),
        }
    })
    .build()
    .await?;

let handle = proxy.run().await?;
handle.shutdown().await?;
```

--------------------------------

### Skill File Structure Example

Source: https://github.com/just-every/code/blob/main/code-rs/core/src/skills/assets/samples/skill-creator/SKILL.md

Illustrates the basic directory structure for a skill, including the required SKILL.md file and optional resource directories.

```text
skill-name/
  SKILL.md
  scripts/     (optional)
  references/  (optional)
  assets/      (optional)
```

--------------------------------

### Bootstrap Flow for New Remote Machines

Source: https://github.com/just-every/code/blob/main/codex-rs/app-server-daemon/README.md

This sequence installs the standalone managed version of Codex and then bootstraps the app server daemon with remote control enabled. The bootstrap process records settings, starts the app server as a detached process, and launches an updater loop.

```sh
curl -fsSL https://chatgpt.com/codex/install.sh | sh
$HOME/.codex/packages/standalone/current/codex app-server daemon bootstrap --remote-control
```

--------------------------------

### Install Codex CLI

Source: https://github.com/just-every/code/blob/main/code-rs/README.md

Install the Codex CLI globally using npm. This is the easiest method for installation.

```shell
npm i -g @openai/codex
codex
```

--------------------------------

### Build and Run App Server Test Client

Source: https://github.com/just-every/code/blob/main/codex-rs/app-server-test-client/README.md

Build the codex binary, start the websocket app-server in the background, and then call the app-server with a model list command.

```bash
# 1) Build debug codex binary
cargo build -p codex-cli --bin codex

# 2) Start websocket app-server in background
cargo run -p codex-app-server-test-client -- \
  --codex-bin ./target/debug/codex \
  serve --listen ws://127.0.0.1:4222 --kill

# 3) Call app-server (defaults to ws://127.0.0.1:4222)
cargo run -p codex-app-server-test-client -- model-list
```

--------------------------------

### Install Companion AI CLIs

Source: https://github.com/just-every/code/blob/main/README.md

Installs Node.js 20+, sets up environment variables for npm global packages, and installs companion AI CLIs like Claude, Gemini, and Qwen. Includes smoke tests to verify installations.

```bash
# Ensure Node.js 20+ is available locally (installs into ~/.n)
npm install -g n
export N_PREFIX="$HOME/.n"
export PATH="$N_PREFIX/bin:$PATH"
n 20.18.1

# Install the 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
curl -fsSL https://antigravity.google/cli/install.sh | bash

# Quick smoke tests
claude --version
agy --version
gemini --version
qwen --version
```

--------------------------------

### Install Skill from GitHub URL

Source: https://github.com/just-every/code/blob/main/code-rs/core/src/skills/assets/samples/skill-installer/SKILL.md

Installs a skill from a specific URL on GitHub. This method is useful for installing skills from branches or specific commits.

```python
scripts/install-skill-from-github.py --url https://github.com/<owner>/<repo>/tree/<ref>/<path>
```

--------------------------------

### MCP Server Configuration Example

Source: https://github.com/just-every/code/blob/main/README.md

Configure Model Context Protocol (MCP) servers in `~/.code/config.toml`. This example shows how to set up a filesystem server using npx.

```toml
[mcp_servers.filesystem]
command = "npx"
args = ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/project"]

```

--------------------------------

### Create and Update Task Plan with update_plan

Source: https://github.com/just-every/code/blob/main/code-rs/core/gpt_5_2_prompt.md

Shows how to initialize a task plan with pending steps and how to update the status of steps to 'in_progress' or 'completed'.

```json
[
  {"step": "Write code", "status": "pending"},
  {"step": "Test code", "status": "pending"},
  {"step": "Deploy code", "status": "pending"}
]
```

```json
[
  {"step": "Write code", "status": "in_progress"},
  {"step": "Test code", "status": "pending"},
  {"step": "Deploy code", "status": "pending"}
]
```

```json
[
  {"step": "Write code", "status": "completed"},
  {"step": "Test code", "status": "in_progress"},
  {"step": "Deploy code", "status": "pending"}
]
```

```json
[
  {"step": "Write code", "status": "completed"},
  {"step": "Test code", "status": "completed"},
  {"step": "Deploy code", "status": "completed"}
]
```

--------------------------------

### Start Process Response

Source: https://github.com/just-every/code/blob/main/codex-rs/exec-server/README.md

Confirms the process has been started, returning its assigned ID.

```json
{
  "processId": "proc-1"
}
```

--------------------------------

### Create, Write, and Read Filesystem Entries

Source: https://github.com/just-every/code/blob/main/codex-rs/app-server/README.md

Demonstrates creating a directory, writing data to a file, and reading data from a file using JSON-RPC methods. Ensure paths are absolute.

```json
{ "method": "fs/createDirectory", "id": 40, "params": {
    "path": "/tmp/example/nested",
    "recursive": true
} }
{ "id": 40, "result": {} }
{ "method": "fs/writeFile", "id": 41, "params": {
    "path": "/tmp/example/nested/note.txt",
    "dataBase64": "aGVsbG8="
} }
{ "id": 41, "result": {} }
{ "method": "fs/readFile", "id": 43, "params": {
    "path": "/tmp/example/nested/note.txt"
} }
{ "id": 43, "result": {
    "dataBase64": "aGVsbG8="
} }
```

--------------------------------

### Structured JSON output example

Source: https://github.com/just-every/code/blob/main/docs/exec.md

Example of the JSON output when using `--output-schema`.

```json
{"project_name":"Every Code CLI","programming_languages":["Rust","TypeScript","Shell"]}
```

--------------------------------

### Install @just-every/code

Source: https://github.com/just-every/code/blob/main/docs/release-notes/RELEASE_NOTES.md

Installs the latest version of the @just-every/code package globally and then runs the code command.

```sh
npm install -g @just-every/code@latest
code
```

--------------------------------

### Build Bubblewrap from Source

Source: https://github.com/just-every/code/blob/main/codex-rs/vendor/bubblewrap/README.md

Instructions for building Bubblewrap using Meson. Ensure you have Meson installed for this process.

```sh
meson _builddir
meson compile -C _builddir
Meson test -C _builddir
Meson install -C _builddir
```