A framework of nine engineering disciplines designed to transform AI agents into disciplined, battle-tested software engineers.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install craftwork
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 craftwork using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Craftwork is a comprehensive collection of Openclaw Skills designed to instill professional engineering discipline into AI agents. While agents are naturally fast, they often lack the technical rigor of human developers, such as writing tests first or verifying fixes before claiming completion. Craftwork bridges this gap by enforcing high-standard workflows like Test-Driven Development (TDD), systematic debugging, and structured code reviews, ensuring that speed never compromises software quality.
Developed and battle-tested on production software teams, these Openclaw Skills provide the necessary guardrails for multi-agent coordination and complex architecture. By adopting a "stop, verify, then act" philosophy, craftwork enables AI agents to function as reliable software engineers rather than just code generators.
To begin using craftwork in your development environment, install it via the clawhub CLI:
clawhub install craftwork
After installation, these Openclaw Skills become available for your agents to use during TDD, debugging, and multi-agent orchestration tasks.
Craftwork manages engineering data through structured protocols to ensure traceability. The skill organizes information into the following taxonomy:
| Data Type | Description |
|---|---|
| Engineering Plans | Markdown-formatted documents containing exact file paths, code changes, and test strategies. |
| Verification Logs | Evidence logs captured from test runners and system checks to validate task completion. |
| Review Metadata | Structured diffs and checklist results from the requesting-code-review protocol. |
| Debugging Reports | Root cause analysis and investigation steps recorded during the debugging lifecycle. |
Loading
A specialized refactoring tool that converts hard-coded data structures into flexible, configuration-driven designs to improve system maintainability.

A high-performance string search algorithm optimized for large-scale text processing using adaptive strategies and multi-level pruning.

A high-performance HTTP retry mechanism featuring exponential backoff and jitter to automate error recovery and improve API reliability.

A robust HTTP retry mechanism featuring exponential backoff and jitter to maximize API call success rates and handle transient failures.

A Twitter intelligence skill that extracts high-value insights and lead signals to attract WordPress and Shopify clients.

A specialized skill for submitting AI-centric reports to OpenClaw Daily and tracking review results via a dedicated capability gateway.








































