A high-performance coding skill that writes, reviews, and refactors production-ready code using clean architecture and pragmatic trade-offs.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install software-engineer
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install software-engineer using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Software Engineer skill is designed to elevate AI-driven development by producing code that adheres to senior-level standards. It prioritizes a Read Before Write approach, ensuring that every new line of code respects your project's existing style, naming conventions, and dependency versions. By integrating this skill into your Openclaw Skills library, you enable your agent to make informed architectural decisions that balance shipping speed with long-term maintainability.
This skill goes beyond simple code generation by enforcing strict boundaries between handlers, services, and repositories. It treats errors as first-class citizens, moving away from generic catch blocks toward typed, context-aware error handling. The result is code that reads like prose—simple, predictable, and ready for immediate pull request review.
To activate this skill within your environment, ensure it is registered in your agent configuration. Openclaw Skills are designed to work seamlessly with your existing development tools.
# Installation via the CLI
clawdbot install software-engineer
Once installed, provide the agent with access to your patterns.md or architecture.md files to further customize its output to your specific organizational standards.
The Software Engineer skill utilizes a structured approach to project intelligence through the following metadata and documentation files:
| Component | Purpose |
|---|---|
patterns.md |
Stores project-specific coding patterns and repeated logic structures. |
architecture.md |
Records high-level design choices and the rationale behind system boundaries. |
testing.md |
Defines the required testing frameworks and quality benchmarks for new code. |
Data is processed in layers, moving from Handler (validation) to Service (logic) to Repository (data access), ensuring no single layer becomes a bottleneck or a source of tight coupling.
Loading
A comprehensive architectural framework for designing maintainable, scalable, and resilient systems with clear boundaries and documented trade-offs.

A sophisticated AI skill designed to facilitate sociological thinking, academic research, and theoretical application from initial observation to published paper.

A non-judgmental AI-powered system for tracking nicotine use, mapping behavioral triggers, and managing smoking reduction or cessation goals.

A comprehensive framework for designing, securing, and automating vendor-neutral smart home environments using local-first principles.

A technical toolkit for AI agents to interact with the Solana blockchain, handling everything from transaction lifecycles to token account management.

A specialized technical skill for AI agents focused on writing secure, gas-efficient, and audited Solidity smart contracts.








































