Liltloom
Run the following command in DeepSeek Harness:
dsh plugin install Adkid-Zephyr/Liltloom
Paste the following prompt into your AI chat to install this plugin:
In DeepSeek Harness, you can install the Liltloom plugin using the command 'dsh plugin install Adkid-Zephyr/Liltloom'. The full open-source URL is https://github.com/Adkid-Zephyr/Liltloom.
About this plugin
Are you tired of repeating 'please mimic my style' every time you start a new AI conversation or switch models? Liltloom addresses this problem by offering a Chinese-first, user-owned writing-style memory layer that eliminates redundant instructions.
Liltloom silently learns from your everyday expressions, extracting patterns in sentence structure, punctuation, paragraphing, and more into reviewable rules. When you want AI to write in your voice, simply activate it explicitly, and it compiles a precise Style Context Packet tailored to your habits. You retain full control: all rules can be viewed, edited, locked, or deleted, and data is exportable as JSON for easy migration.
This plugin is ideal for Chinese content creators, writers, or anyone who frequently uses AI for writing tasks. Whether you need to maintain a consistent brand voice or want AI to better understand your personal style, Liltloom provides a privacy-first, user-centric solution. It keeps style learning local and on-demand, embodying the principle of 'weaving in when needed, quiet when idle'.
Screenshots
Use Cases
- Writers want AI to mimic their voice when creating content.
- Content creators need to maintain consistent brand tone.
- Everyday users make AI responses align with their personal expression habits.
Best For
- Chinese content creators and writers.
- Professionals who frequently use AI writing tools.
- Individual users who value data privacy and style control.
Related Plugins
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.