A persistent memory skill that captures and applies your specific coding preferences and stack decisions through explicit user feedback.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install taizi-coding
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install taizi-coding using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Coding?
The Coding skill is a specialized utility within the ecosystem of Openclaw Skills designed to provide your AI agent with a long-term memory of your development standards. Unlike automated tools that might incorrectly infer style from a project, this skill operates on a high-trust model, only learning from your direct corrections and explicit confirmations. It ensures that every line of code generated reflects your preferred frameworks, naming conventions, and architectural patterns, making the AI feel like a true extension of your own development workflow.
By centralizing your preferences in local Markdown files, this skill provides a privacy-first approach to personalization. It bridges the gap between different coding sessions, ensuring that once you correct a specific pattern, the AI remembers it forever. This makes it an essential component for developers using Openclaw Skills who demand consistency and precision in their automated coding tasks.
Coding Use Cases
- Standardizing naming conventions such as snake_case for files or camelCase for variables across all sessions.
- Enforcing specific tech stack decisions, such as requiring Python 3.11+ or specific React versions.
- Maintaining consistent project structures, including test colocation and folder hierarchies.
- Creating a blacklist of rejected libraries or patterns that the agent should never suggest.
How Coding Works
- When a user provides a correction to a code snippet, the agent identifies a potential preference.
- The agent asks the user for explicit confirmation to store this preference within the Openclaw Skills directory.
- Confirmed preferences are converted into ultra-compact, five-word entries and appended to a local memory file.
- At the start of every new coding session, the agent automatically loads the existing preferences to set the context.
- If the active memory exceeds 100 lines, the skill intelligently archives older entries to a history file to prevent context bloat.
Coding Setup
To initialize this skill among your local Openclaw Skills, you must first create the required directory structure for data persistence:
mkdir -p ~/coding
Once the directory is created, the skill will automatically generate the memory.md and history.md files upon your first confirmed preference.
Coding Data Schema & Taxonomy
The skill manages its data within the ~/coding/ directory using a tiered Markdown approach to ensure Openclaw Skills remain lightweight and searchable:
| File | Role | Description |
|---|---|---|
memory.md |
Active Context | Stores up to 100 lines of current, high-priority preferences. |
history.md |
Archive | Contains older preferences that have been rotated out of active memory. |
dimensions.md |
Taxonomy | Defines categories like Stack, Style, Structure, and Never. |
criteria.md |
Logic | Details the specific conditions required for a preference to be stored. |
Coding Advanced Features
- Smart memory rotation that automatically moves entries to history.md to maintain optimal performance and context window management.
- Category-based organization using the Dimensions framework to separate technology stacks from stylistic choices.
- Natural language query support, allowing you to ask the agent to show, forget, or explain specific stored preferences.
- Local-only data storage ensures your proprietary coding patterns and preferences never leave your machine.
SKILL.md
Loading
Related Openclaw Skills
Project Context Manager
A project-based context management system that enforces structured documentation and persistent memory for AI coding agents.
Porteden Docs
A secure CLI-driven tool for comprehensive Google Docs lifecycle management and content manipulation optimized for AI agents.
Video Knowledge Ingest
A powerful cross-platform tool to transform video URLs or local files into searchable, summarized local knowledge using Whisper and yt-dlp.
Novel
A lightning-fast terminal utility for managing novel metadata, including chapters, characters, and plot timelines without leaving the command line.
Beijing Morning Weather Assistant
An automated morning briefing skill that provides Beijing weather updates along with specific lifestyle recommendations for clothing and nutrition.
AI Intelligent Contract Review
Automate contract review, risk identification, and clause analysis with high-precision AI models.
Featured*
