wuzhanwei
Community@vincent-ng
https://vincentng.cn/
Agent Skills by wuzhanwei
Showing 20 vetted skills indexed across 1 GitHub repositories.
using-git-worktrees
Create isolated git worktrees with directory detection and ignore validation.
test-driven-development
Enforces red-green-refactor cycles for software development workflows.
systematic-debugging
Guide engineers through a four-phase root-cause investigation before proposing fixes.
chinese-git-workflow
Configure Git workflows for Gitee, Coding, and 极狐 GitLab.
using-superpowers
Invoke the Skill tool before every assistant response to enforce pre-response checks.
dispatching-parallel-agents
Dispatch independent agents to parallel tasks and aggregate results.
chinese-code-review
Generate balanced Chinese-language code review feedback with actionable improvements.
executing-plans
Coordinates end-to-end plan execution with checkpoints and traceable task updates.
finishing-a-development-branch
Validate tests and apply a merge, PR, or cleanup workflow for a development branch.
chinese-documentation
Standardize spacing, punctuation, and terminology in Chinese technical documents.
brainstorming
Guide structured brainstorming to transform ideas into design specifications.
workflow-runner
Execute agency-orchestrator YAML workflows with role emulation and variable substitution.
writing-plans
Translate vague specifications into concrete, testable implementation plans.
requesting-code-review
Dispatch a code-reviewer sub-agent with base/head SHAs and a review template.
receiving-code-review
Clarify and validate code-review feedback against the codebase and standards.
writing-skills
Standardize reusable agent skill writing with TDD-driven validation.
chinese-commit-conventions
Standardize Chinese-language commit messages with commitlint and husky.
mcp-builder
Automate construction of MCP servers in TypeScript and Python.
verification-before-completion
Enforce evidence-based validation of completion claims during pull requests and releases.
subagent-driven-development
Orchestrate independent subagents executing per-task plans with two-stage reviews.