mcpField
O
omniroute-integration

omniroute-integration

github.com/nianyi778Agent workflows143 installsUpdated 33w ago
Repository

About omniroute-integration

omniroute-integration is a skill that configures OpenCode to use the OmniRoute AI gateway. It is intended for OpenCode users who want to access and switch among models from multiple providers through one configuration.

Key features

  1. 01Configure the OmniRoute provider in OpenCode
  2. 02Add available models to OpenCode settings
  3. 03Save OmniRoute authentication credentials
  4. 04Set a default model
  5. 05Test the connection with an API call

Use cases

  1. 01Connect OpenCode to the OmniRoute gateway
  2. 02Configure models from Qwen, Codex, GitHub, and Kiro
  3. 03Switch between models available through OmniRoute

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/nianyi778/opencode-omniroute-integration --skill omniroute-integration

After installing, type/omniroute-integrationin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

LICENSE
README_ZH.md
.gitignore
create_github_repo.sh
install.sh
README.md
example_usage.md
package.json
PUBLISH_GUIDE.md
skill.yml

Details

From the project README

OmniRoute Integration Skill

Configures OpenCode to work with the OmniRoute AI gateway service, enabling access to multiple AI models from various providers including Qwen, Codex, GitHub and Kiro.

When to use

Use this skill when you want to:

  • Connect OpenCode to the OmniRoute AI gateway
  • Access multiple AI models through a single interface
  • Configure all available OmniRoute models in OpenCode
  • Switch between different AI providers easily

Prerequisites

Instructions

  1. Obtain your OmniRoute API key
  2. Run the skill to initiate configuration
  3. Enter your API key when prompted
  4. Confirm the OmniRoute service endpoint (default: http://138.2.29.190:20128)
  5. Allow the script to configure OpenCode settings
  6. Verify configuration by testing a model

Configuration Details

The skill will:

  • Create/update ~/.config/opencode/opencode.json with OmniRoute provider
  • Add all available models (Qwen, Codex, GitHub, Kiro) to the configuration
  • Update authentication credentials in ~/.local/share/opencode/auth.json
  • Set up default model omniroute/qw/qwen3-coder-plus

Available Models

After successful configuration, you can use models such as:

  • omniroute/kr/claude-sonnet-4.5
  • omniroute/kr/claude-haiku-4.5
  • omniroute/qw/qwen3-coder-plus
  • omniroute/qw/qwen3-coder-flash
  • omniroute/cx/gpt-5.2-codex
  • omniroute/gh/gpt-4o
  • And over 100 other models available through OmniRoute

Verification

The skill will automatically test the configuration with a simple API call to verify the connection works properly.

Comments

Markdown supported