Code Quality Workflow

End-to-end code quality workflow for evidence-backed code review, full repository reviews, architecture reviews of software and character-card systems, refactor decisions, authorized minimal changes, and verification. Use for 全量审查、架构审查、全量架构审查, code or diff audits, cleanup triage, or a scoped quality fix. Preserve coverage gaps and keep audit-only requests read-only. Soul/persona voice does not change review ownership.

224 stars 20 forksSoftware Development Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add LiarMTTT/TavernWeave

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add LiarMTTT/TavernWeave --skill code-quality-workflow
Manual install
git clone --depth 1 https://github.com/LiarMTTT/TavernWeave.git /tmp/TavernWeave
mkdir -p ~/.claude/skills
cp -r /tmp/TavernWeave/skills/code-quality-workflow ~/.claude/skills/code-quality-workflow

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Code Quality Workflow

## Shared communication

Apply [TW plain-language and guidance rules](../consult-tavernweave-library/references/communication-and-guidance.md): explain terms on first use, keep the user-selected 新人/入门/熟练/老手 level, and still explain needed changes and bugs at every level. Soul and prose modes never disable this baseline.

## Plugin runtime

When this Skill runs through a plugin, cloud host, or remote machine, read [the plugin execution contract](../consult-tavernweave-library/references/plugin-runtime.md) at entry and apply its host, tool, and persistence checks before host-specific commands. Keep this Skill's scope, authorization, and acceptance gates.

Before an authorized write, use `$consult-tavernweave-library` with the `code-quality-workflow` route and record the A0 receipt. When the last acceptance gap keeps expanding, read [references/finish-mode.md](references/finish-mode.md) and freeze one visible problem plus one exit condition.

Use one state machine for code-quality work:

```text
AUDIT -> SWEEP -> GATE -> CHANGE -> VERIFY
  |        |        |        |         |
  +--------+--------+--------+---------+--> REPORT / STOP
```

Skip `SWEEP` for a focused target. Stop after any read-only state when the request is only for analysis. Never treat a cleanup request as permission to edit.

## Non-negotiable rules

1. Prefer the simplest maintainable solution, the fewest changed files, the fewest new abstractions, and the fewest dependencies.
Read the full skill on GitHub

Details

Skill name
code-quality-workflow
Path
skills/code-quality-workflow/SKILL.md
Repo updated
4 November 2026

FAQ

What does the Code Quality Workflow skill do?

End-to-end code quality workflow for evidence-backed code review, full repository reviews, architecture reviews of software and character-card systems, refactor decisions, authorized minimal changes, and verification. Use for 全量审查、架构审查、全量架构审查, code or diff audits, cleanup triage, or a scoped quality fix. Preserve coverage gaps and keep audit-only requests read-only. Soul/persona voice does not change review ownership.

How do I install the Code Quality Workflow skill?

Claude Code plugin: /plugin marketplace add LiarMTTT/TavernWeave. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Code Quality Workflow popular?

It lives in LiarMTTT/TavernWeave, which has 224 GitHub stars and 20 forks.

Related skills

Activate Tavernweave Soul

224

LiarMTTT/TavernWeave

Activate, switch, combine, or close TavernWeave Soul as a portable current-task teaching overlay with 阿瞳 for warm guidance, MTTT.sir for strict learning examination, 灵魂杀手 for blunt frontend commentary through a disclosed Johnny Silverhand fan-roleplay Easter egg, and a three-seat ensemble of persona lenses for bounded brainstorming. Use when the user directly invokes a supported Soul command, says “脑暴模式,Soul 联席”, or asks for these guidance modes while making cards, reviewing frontend, or Vibe Coding. Soul provides persona, voice, and review lenses only: full/architecture review is owned by `$c

Frontend & Design20

Build Work Library

224

LiarMTTT/TavernWeave

Build source-backed work reference libraries for fan creation from novels, local material, researched webpages and inspected image references. Use for 小说转资料库/数据库、作品资料库、同人资料整理、联网收集文图、原作设定检索与二创资料包. Preserve canon/fanon, adaptation, spoiler stage and provenance; produce editable Markdown, rebuildable JSONL and optional SQLite. Not the TW bundled design catalog, a generic SQL app, or an in-card MVU database.

Software Development20

Consult Tavernweave Library

224

LiarMTTT/TavernWeave

Route TavernWeave work to bundled development guides, the standing A0 checklist, screened design and motion catalogs, concept/source references, and the offline picker. Also provide TW plain-language guidance and explicit user-selected 新人/入门/熟练/老手 preference management. Use before TW file writes, for relevant domain or design guidance, or when the user asks to choose, query or change an onboarding level. Do not load the entire library, publish excluded private material, infer a user level, or treat a catalog candidate as adopted work.

Software Development20

Reflect on Vibe Code Growth

224

LiarMTTT/TavernWeave

Reassess a user's complete Vibe Code capability network from currently available host conversations, workspace, project-state, and delivery evidence; produce both a detailed standalone text assessment form and a seven-part interactive Vibe Code 成长历程 report; compare only genuinely comparable dimensions with prior immutable mirror records; and optionally save an append-only local history. Use when the user says 照镜子, 分析我的 Vibe Code 成长历程, 再次核验成长, 和上次履历比较, 保存照镜子履历, 从上一次履历继续分析, 只要文字评估, or 导出文字评估表单. Keep exact measurements, user reports, estimates, behavior-evidence scores, and unverifiable claims se

Software Development20

API Triage

272

xigua-wang/skill-doctor

Investigates API integration failures and remote behavior.

Software Development

Sillytavern API Reference

224

LiarMTTT/TavernWeave

Verify exact SillyTavern, Tavern Helper / JS-Slash-Runner, STScript, macro, prompt-injection, worldbook, EJS, MVU, and runtime-library capabilities before implementing or reviewing rolecard automation. Use when a task depends on a function signature, event payload, slash-command grammar, variable scope, message-floor operation, generation option, runtime-symbol provider, Git/CDN loader, or version-sensitive extension capability.

Software Development20

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.