wangbinquan

wangbinquan/agent-workflow

Instructions file CodexOpenCode

Instructions for wangbinquan/agent-workflow, covering repository agent instructions and commit attribution.

wangbinquan/agent-workflow

Instructions file

A set of instructions for an AI coding agent working in a software repository. It explains how to track repository status, propose work through RFCs (written design documents), collaborate on changes, and add tests.

wangbinquan/agent-workflow

Agent

A read-only audit agent that waits for all audit parts to finish, then combines their findings into one report. It removes duplicates, groups issues by priority and file, and keeps supporting evidence and recommendations.

wangbinquan/skills

Agent Claude Code

An AI agent that summarizes every change in the latest Git commit, a saved set of repository changes, in Chinese. It uses the original requirements or design document for context and saves the summary beside it.

wangbinquan/skills

Agent Claude Code

Um agente que compara documentos de design com o código para verificar se a implementação corresponde ao que foi planejado.

wangbinquan/skills

Agent Claude Code

An AI agent that turns a requirements document into a detailed software design document based on the current codebase. The design can guide code generation, test-driven development (writing tests before implementation), and implementation.

wangbinquan/skills

Agent Claude Code

A coding agent for the later implementation stage of test-driven development, or TDD, where tests and a code skeleton guide the completion of business logic. It fills existing TODO sections and does not generate the skeleton or write unit tests.

wangbinquan/skills

Agent Claude Code

A specialised assistant for generating empty code structures from requirements or design documents as the first step of TDD, or test-driven development.

wangbinquan/skills

Agent Claude Code

A dedicated coding agent for generating unit tests, which are small automated checks for individual functions or classes.

wangbinquan/skills

Skill Claude CodeCodex

A skill for reviewing, completing, and adding business-system logs. Logs are records of application events used to understand behavior, diagnose failures, and support audits.

wangbinquan/skills

Skill Claude CodeCodex

A code-review skill that examines a repository from an SRE perspective. SRE, or site reliability engineering, focuses on keeping software dependable in production.

wangbinquan/skills

Skill Claude CodeCodex

A review skill that compares design documents with the code built from them. It checks whether the implementation matches the stated design, specification, or contract.

wangbinquan/skills

Skill Claude CodeCodex

A skill for turning lessons from debugging, incidents, code reviews, load tests, and project retrospectives into reusable engineering knowledge stored in the code repository.

wangbinquan/skills

Skill Claude CodeCodex

A planning and execution method for generating large amounts of code from design documents, such as API, database, architecture, or interface specifications. It divides the work among subagents, which are separate AI workers.

wangbinquan/skills

Skill Claude CodeCodex

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

wangbinquan/skills

Agent Claude Code

Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.

wangbinquan/skills

Agent Claude Code

Compare two outputs WITHOUT knowing which skill produced them.

wangbinquan/skills

Agent Claude Code

Evaluate expectations against an execution transcript and outputs.

wangbinquan/skills

Skill Claude CodeCodex

Draw software-related diagrams (architecture, sequence, class, flowchart, state machine, ER, component, deployment, activity, use-case, data-flow, gantt, package, deployment-topology, etc.) using either Mermaid or PlantUML. Every PlantUML diagram gets Chinese (or mixed CJK) note annotations for extra context, every…

wangbinquan/skills

Skill Claude CodeCodex

A skill that creates an initial code structure from requirements or design documents before tests are written. TDD, or test-driven development, is a method where tests guide the implementation.

wangbinquan/skills

Skill Claude CodeCodex

A test-driven development implementation tool. TDD means writing tests around expected behavior and then implementing code that satisfies them.

wangbinquan/skills

Skill Claude CodeCodex

A skill that designs and generates unit tests for Java, C++, Python, Go, Rust, Scala, JavaScript, and TypeScript. Unit tests check small parts of a program independently.

wangbinquan/skills

Agent

../../demo/agents/task-completion-checker.md.

wangbinquan/skills

Instructions file

Claude Code instructions for wangbinquan/skills, covering claude.md, what this repo is, top-level layout, the skill/agent pipeline (how the pieces compose) and naming convention for generated artifacts.