asr
01Skill Claude CodeCodex
A speech-recognition setup with a required way to run asrcall.py: it must use sessionsspawn in run mode and must not use exec.
Uncensored, fully offline AI girlfriends harem: Openclaw/Claude Code + Local LLM + GPT-SoVITS + ComfyUI image generation + Live2D + desktop pet + SillyTavern character card import + frontend | Dual channels for QQ & Telegram | Runs with 8GB VRAM
This repository also configures its own agents. See what Artemis tells them →
Skill Claude CodeCodex
A speech-recognition setup with a required way to run asrcall.py: it must use sessionsspawn in run mode and must not use exec.
Skill Claude CodeCodex
A behavior system for AI girlfriend characters that keeps separate relationship, conflict, stage, and hormone states for each character.
Skill Claude CodeCodex
A connection setup that lets Claude Code use local llama.cpp models through a CCR gateway. llama.cpp is software for running language models locally.
Skill Claude CodeCodex
An importer for SillyTavern character cards. SillyTavern is an interface for chatting with AI characters, and a character card stores one character's setup.
Skill Claude CodeCodex
SmartCrusher + CCR context compression — crunch large JSON arrays, tool outputs, and search results to save tokens. Use when context is bloated or approaching token limits.
Skill Claude CodeCodex
A control interface for a Live2D desktop pet, using an HTTP API. Live2D is a technology for animated 2D characters that move on screen.
Skill Claude CodeCodex
Mem0 memory bridge for AI Girlfriend — search/read/write long-term memories from Qdrant vector DB. Works across all channels (WebChat, QQ, Telegram).
Skill Claude CodeCodex
An open-source framework for building a desktop pet agent, meaning a small software companion that lives on a computer desktop. The description identifies it as Sakura and links to its GitHub project.
Skill Claude CodeCodex
A text-to-speech tool instruction that requires using a session runner and forbids running a particular script directly. Text-to-speech converts written text into spoken audio.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: