A local memory system that ensures your AI agent follows your specific coding standards and technical preferences through explicit feedback.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install coding-1-0-3
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 coding-1-0-3 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 Skill for AI Agents?
The Coding skill is a specialized memory module designed for developers who want their AI agents to adhere to a specific set of stylistic and architectural rules. Unlike generic agents that guess patterns from your existing files, this component of Openclaw Skills only learns from your direct corrections and confirmations. This approach ensures a high level of precision in code generation, maintaining consistency across projects without the need for repetitive prompting or manual configuration files in every directory.
Coding Skill for AI Agents Use Cases
- Enforcing specific naming conventions such as snake_case or PascalCase for specific file types.
- Storing preferred framework versions and library choices to avoid outdated code suggestions.
- Defining structural patterns like the location of test files or configuration management.
- Maintaining a list of prohibited patterns or libraries that the agent should never suggest.
How Coding Skill for AI Agents Works
- Provide an explicit correction when the agent generates code that doesn't match your preference.
- Confirm the agent's request to store that specific preference in your local Openclaw Skills memory.
- The agent writes a compact, five-word entry to the memory.md file located in your home directory.
- Future sessions automatically load these preferences to tailor the agent's output to your established style.
- Manage your preferences by querying the agent to show, forget, or update specific entries in the local database.
Coding Skill for AI Agents Setup
To initialize the environment for this skill, run the following command to create the storage directory:
mkdir -p ~/coding
The skill will manage the creation of memory.md and other reference files during its first interaction. Ensure your environment is synced by using clawhub sync to get the latest updates for Openclaw Skills.
Coding Skill for AI Agents Data Schema & Taxonomy
The skill organizes data within the ~/coding/ directory using a clear Markdown-based taxonomy:
| File | Role | Description |
|---|---|---|
| memory.md | Primary Store | Contains up to 100 lines of active, ultra-compact coding preferences. |
| history.md | Archive | Stores older preferences that have been moved out of active context. |
| dimensions.md | Schema | Defines categories such as Stack, Style, Structure, and Never. |
| criteria.md | Policy | Outlines the logical triggers for when a preference should be recorded. |
Coding Skill for AI Agents Advanced Features
- Context-aware preference application that loads relevant style rules at the start of every session.
- Automatic entry merging to consolidate similar preferences and reduce context window usage.
- Privacy-first architecture that stores all developer preferences locally without cloud synchronization.
- Support for granular style queries allowing users to inspect what the agent knows about specific languages or stacks.
SKILL.md
Loading
Related Openclaw Skills
Slack Integration
Enable your AI agent to manage Slack workspace communication, reactions, and pinned messages using official bot tokens.
Trello Skill
A developer-centric skill for programmatically managing Trello boards, lists, and cards via the REST API.
Solana Blockchain & Wallet Manager
A comprehensive toolset for interacting with the Solana blockchain, managing assets, and executing DeFi transactions via Helius infrastructure.
User Insight
An advanced skill for building dynamic user profiles through passive information extraction and proactive topic exploration to optimize communication.
DingTalk Docs
A comprehensive automation skill for managing documents, folders, and content within the DingTalk cloud ecosystem.
DingTalk Channel Plugin
A powerful enterprise messaging integration for Openclaw Skills that enables seamless communication through DingTalk.
Featured*
