qa-skills
Run the following command in DeepSeek Harness:
dsh plugin install fishzjp/qa-skills
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, please run the command `dsh plugin install fishzjp/qa-skills`, and you can access the full source code at https://github.com/fishzjp/qa-skills .
About this plugin
AI-assisted testing often falls into the trap of generating professional-looking but unexecutable test cases filled with vague assertions and placeholders, while lacking genuine strategic decision-making. qa-skills is built to solve this pain point, empowering AI agents to think and work like senior QA engineers through a structured instruction framework.
Driven by a Knowledge × Tools × Decision architecture, this framework orchestrates a comprehensive 9-stage testing pipeline. It introduces a rigorous Type Decision Matrix and an Evidence & Risk Model to ensure every strategy is auditable and comprehensively scoped. With a file-as-state design, it overcomes long-context limitations and supports interrupted workflows, ultimately delivering both human-readable mind maps and machine-executable assets.
Whether you are a developer leveraging AI coding agents like Claude Code, Cursor, or DeepSeek Harness, or a QA team aiming to deeply integrate AI into testing workflows, qa-skills helps you build highly executable, end-to-end traceable automated testing pipelines, unlocking AI's true potential in quality assurance.
Screenshots
Use Cases
- Automatically generate executable test cases and mind maps from requirement documents
- Analyze code change risks and intelligently formulate end-to-end testing strategies
- Convert existing manual test cases into Playwright automation scripts
Best For
- Software engineers using AI agents like Claude Code or Cursor
- QA engineers looking to improve test case writing and review efficiency with AI
- Technical leads dedicated to building AI-driven automated testing pipelines
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.