rai-sds

Rendered.ai Synthetic Data Studio CLI

A fast, minimal-dependency CLI for managing synthetic datasets, graphs, channels, workspaces, volumes, and more on the Rendered.ai Synthetic Data Studio platform.

Claude Code | Gemini CLI | Codex | CLI Only | CLI Usage

AI Agent Integration

The rai-sds skill uses the Agent Skills open standard and works across Claude Code, Google Gemini CLI, and OpenAI Codex.

Claude Code

Plugin (recommended — auto-installs binary + skill):

# 1. Add the marketplace (one time)
/plugin marketplace add https://github.com/renderedai/rai-synthetic-data-studio-cli

# 2. Install the plugin
/plugin install rai-sds

Or use /plugin and follow the interactive menu.

This installs the rai-sds plugin, which:

• Adds the /rai-sds:rai-sds skill to Claude Code
• Adds the /rai-sds:setup command for guided onboarding

After installing, reload plugins to activate the new commands (no restart needed):

/reload-plugins

Then run /setup to get started:

/rai-sds:setup

This will check if you're logged in, walk you through authentication (or registration), and help you set your organization and workspace context.

Standalone skill (skill only):

curl -fsSL https://raw.githubusercontent.com/renderedai/rai-synthetic-data-studio-cli/main/install-skill.sh | bash

Installs the binary and skill to ~/.claude/skills/rai-sds/. Run /reload-plugins to activate /rai-sds without restarting.

---

Google Gemini CLI

Extension (recommended — includes context file + skill):

gemini extensions install https://github.com/renderedai/rai-synthetic-data-studio-cli

Standalone skill (skill only):

curl -fsSL https://raw.githubusercontent.com/renderedai/rai-synthetic-data-studio-cli/main/install-gemini-skill.sh | bash

Installs the binary and skill to ~/.gemini/skills/rai-sds/. Restart Gemini CLI to activate.

---

OpenAI Codex

curl -fsSL https://raw.githubusercontent.com/renderedai/rai-synthetic-data-studio-cli/main/install-codex-skill.sh | bash

Installs the binary and skill to ~/.codex/skills/rai-sds/. Restart Codex to activate.

---

CLI Only (Linux/macOS)

For terminal users who want the rai-sds binary without AI agent integration.

One-liner (recommended)

curl -fsSL https://raw.githubusercontent.com/renderedai/rai-synthetic-data-studio-cli/main/install.sh | bash

This downloads the pre-built binary for your platform, verifies its checksum, and installs it to ~/.local/bin/rai-sds.

To install to a custom location:

RENDEREDAI_INSTALL_DIR=/usr/local/bin curl -fsSL https://raw.githubusercontent.com/renderedai/rai-synthetic-data-studio-cli/main/install.sh | bash

Build from source

If the one-liner doesn't work (e.g., restricted network, unsupported platform), you can build from source.

Dependencies:

• git
• Rust/cargo — install with: curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

git clone https://github.com/renderedai/synthetic-data-studio-cli.git
cd synthetic-data-studio-cli
cargo build --release
cp target/release/rai-sds ~/.local/bin/

Quick Start

1. Authenticate

Log in via browser (OAuth2 PKCE flow):

rai-sds auth login

2. Set Your Context

Select your organization and workspace:

rai-sds orgs use
rai-sds workspaces use

3. Explore Your Data

# List channels and graphs
rai-sds channels get --format table
rai-sds graphs get --format table

# Create a dataset from a graph
rai-sds datasets create --graph-id <GRAPH_ID> --name "my-dataset" --runs 10

# Download a dataset
rai-sds datasets download --dataset-id <DS_ID> --extract

# Check system status
rai-sds status

4. Log Out

rai-sds auth logout

Authentication Methods

The CLI supports multiple auth methods, resolved in this priority order:

Priority	Method	Usage
1	Bearer token	--bearer-token <TOKEN> or RENDEREDAI_BEARER_TOKEN env var
2	API key (CLI)	--api-key <KEY> or RENDEREDAI_API_KEY env var
3	API key (config)	Stored in ~/.rai-sds/config.yaml
4	Keychain token	Stored automatically after auth login

Using an API Key

# Via flag
rai-sds datasets get --api-key <YOUR_KEY>

# Via environment variable
export RENDEREDAI_API_KEY=<YOUR_KEY>
rai-sds datasets get

Environments

Name	Flag	API
Production (default)	--env prod	api.rendered.ai
Test	--env test	api.test.rendered.ai
Dev	--env dev	api.dev.rendered.ai

Commands