config-setup

Configure octo providers, models, API keys, and agent defaults via REST API.

88|15|Updated May 25, 2026
One-click install
npx skills add https://github.com/open-octo/octo-agent --skill config-setup-open-octo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: config-setup
Source: https://github.com/open-octo/octo-agent/tree/main/internal/skills/defaults/config-setup
Command: npx skills add https://github.com/open-octo/octo-agent --skill config-setup-open-octo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill turns model, provider, API key, and agent preference changes into a guided configuration workflow, so users can update octo without manually navigating configuration files or guessing API details.

Core Features & Use Cases

  • Provider and Endpoint Setup: Add AI providers, API keys, custom base URLs, protocols, and supported model names through octo's REST API.
  • Agent Defaults: Adjust reasoning effort, permission mode, visible reasoning, coauthor settings, and the default workspace directory.
  • Model Management: Add or remove endpoint models, assign default and lite models, and switch the model for an active session.
  • Safe Configuration Workflow: Read current settings before modifying them, validate changes, confirm destructive actions, avoid exposing saved API keys, and fall back to direct config file editing when the server is unavailable.
  • Use Case: Configure a custom OpenAI-compatible relay with a primary model for normal work and a faster model for lightweight tasks, then make it the default for new sessions.

Quick Start

Ask the config-setup skill to show the current octo configuration and help add or modify a provider, model, or agent setting.

Frequently Asked Questions about config-setup

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I configure octo API keys and add a custom REST API provider?

Configuring octo API keys and custom providers involves adding endpoints, base URLs, and supported model names through octo's local REST API, automatically handling validation and API-key exposure safeguards.

Can I switch the AI model for an active octo session without restarting?

Yes, active-session model switching allows you to change the current AI model directly, assigning a different endpoint model to an ongoing session without requiring a restart.

How do I change agent defaults like reasoning effort and permission mode?

Agent defaults such as reasoning effort, permission mode, and coauthor settings are adjusted through the configuration workflow, applying validated preferences to new agent sessions.

What happens if the local octo server is unavailable when I need to update settings?

When the local octo server is unavailable, the configuration workflow initiates an offline fallback to directly edit the user's octo configuration file, ensuring settings updates persist safely.

Do I need to manually edit configuration files to set up an OpenAI-compatible relay?

No, setting up an OpenAI-compatible relay does not require manual file editing; the workflow handles adding custom base URLs, assigning primary and lite models, and setting defaults via the REST API.