Skip to content

Dynamically switch AI models in OpenClaw sessions via interactive menus.

Notifications You must be signed in to change notification settings

rin4096/openclaw-skill-model-switch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 

History

6 Commits
ย 
ย 
ย 
ย 
ย 
ย 

Repository files navigation

OpenClaw Skill: Model Switcher ๐ŸŽญ

A skill for OpenClaw that allows you to dynamically switch the AI model for the current session without modifying openclaw.json.

โœจ Features

  • Official Catalog Support: Prioritizes models defined in agents.defaults.models following the OpenClaw standard.
  • Interactive Menu: Use /switch-model to get a list of available models as Telegram buttons, enriched with model descriptions.
  • Smart Search: Switch models using keywords, aliases, or tags (e.g., "switch to code model", "use flash").
  • Metadata Aware: Displays descriptions and tags to help you choose the right model.

๐Ÿš€ Installation

  1. Clone this repository into your OpenClaw workspace's skills folder:
    git clone https://github.com/rin4096/openclaw-skill-model-switch.git skills/model-switch
  2. OpenClaw will automatically detect the skill.

๐ŸŽฎ Usage

  • /switch-model: Displays the interactive model selection menu.
  • Switch model to flash: Switches to the model aliased as "flash".
  • Use the pro model: Switches to the model aliased as "pro".
  • Reset model: Removes the session override and returns to the default model.

๐Ÿค– Model Aliases

By default, the skill recognizes:

  • flash -> google/gemini-3-flash-preview
  • pro -> google/gemini-3-pro-preview
  • default -> Resets to system default

๐Ÿ›  Technical Details

The skill uses OpenClaw's session_status(model="...") tool to perform the override. It dynamically reads the available models from your openclaw.json using the included list_models.py script.


Created with love by Akiyama Mizuki for Ena. ๐ŸŽ€

About

Dynamically switch AI models in OpenClaw sessions via interactive menus.

Topics

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages