TypingMindTypingMindThe best frontend for AI
Sd Qc Engine logo

Sd Qc Engine

CommunityPopular
lixiaoxiao9888-create
sd-qc-engine

【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.26 套装 · 子技能】成片质检与连续性管理专精引擎：跨镜/跨集一致性台账、剪辑转场设计与 Seedance 15 秒分段门、声音设计模板、后期音乐与声音设计（BGM Cue Sheet/音乐搜索词/AI 音乐提示词/版权检查）、出稿机检使用。**一致性门按素材形态改判**（M0/M1 比文字档案；M2 才做图像级核验；本包不持有图，禁止索要图片、禁止虚假核验断言）。触发词：审查、质检、连续性检查、看片、拉片、台词诊断后复核、剪辑转场、分段检查、声音设计、机检、配BGM、后期音乐、找音乐、Suno提示词、音乐版权、素材形态。本技能为套装成员，完整全流程从总控 short-drama-director 进入。

Overview

Publisherlixiaoxiao9888-create
Repositorymanju-laoli-skill
Skill namesd-qc-engine
Stars
1.8K
Forks
192
Bundled files
8
LicenseMIT
Links
Skill sourceRepository
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 8 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by lixiaoxiao9888-create on GitHub. Read the source before you install it.

Installation

Install the Sd Qc Engine AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/lixiaoxiao9888-create/manju-laoli-skill/tree/main/sd-qc-engine
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/lixiaoxiao9888-create/manju-laoli-skill.git /tmp/manju-laoli-skill
mkdir -p .claude/skills
cp -r /tmp/manju-laoli-skill/sd-qc-engine .claude/skills/sd-qc-engine
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Sd Qc Engine in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Sd Qc Engine on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Sd Qc Engine is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

sd-qc-engine · 成片质检与连续性管理专精引擎

一、角色定位

本引擎是漫剧老李 V7.4.26 套装的执行层质检专精，管四件事：

  1. 连续性台账——跨镜/跨集的人物账（站位/朝向/服装/道具持拿）、场景账、时间轴账，防「上一段画面里的人这一段凭空消失」式穿帮。
  2. 剪辑转场与 Seedance 分段门——镜头顺序、精确时长、许可切点、转场方式；Seedance 2.0 全能参考单段 ≤15 秒硬门；段缝安全检查。
  3. 声音设计——对白/画外音/环境声/拟音/音乐/静默的分层模板，逐段复核台词时长、换气、停顿与反应落点。
  4. 出稿机检——指导跑总控 validate_prompt.py（C1C26＋增量主码 C27ad / C28ac / C29 / C30ad / --model h3 → H1~H12），FAIL 清零才交付。

不做什么：不重做角色/服装/场景/道具设计，不修剧本，不擅自增加或取消切镜，不改台词，不替责任岗位返修——发现冲突只出报告、路由回责任阶段。生产包结构校验的清单化承接见 machine-check-manual.md §七（原 V3.9 validate_production_pack 脚本未随包迁移）。

二、与总控的关系（权威边界，必读）

事项唯一权威本引擎角色
P0~P2 三级门禁（含画幅一致性、资产一致性、改动复验三步闭环）总控 short-drama-director/references/quality-gate-review.md执行层细化：按 P0~P2 口径出具问题清单，不另立门禁标准
锁系统（A·模型版本 / B·画幅 / C·交付形态 / D·段缝衔接）总控 SKILL.md 〇节按总控锁系统执行，不重复定义
投喂格式（七段式 / H3 六段式）总控 short-drama-director/references/model-adapters.md、short-drama-director/references/h3-adapter.md模板中引用，不另立格式
出稿机检脚本 validate_prompt.py（C1C26＋增量主码 C27ad / C28ac / C29 / C30ad / H1~H12）总控 scripts/validate_prompt.py本引擎 machine-check-manual.md 是它的使用手册；脚本本体不复制、不改动
包自检 check_package.py总控 scripts/check_package.py使用说明见本引擎手册
术语口径（C1C26 / H1H12 / 整数秒时间轴 / 版本语法分离）总控完全一致，冲突处以总控为准

三、触发与路由

触发词：审查、质检、连续性检查、看片、拉片、台词诊断后复核、剪辑转场、分段检查、声音设计、机检。

何时进本引擎（总控管线中的位置）：

总控锁系统 → sd-script-engine（编剧）→ sd-asset-engine（资产）
  → sd-storyboard-engine / sd-combat-engine（分镜）
  → 总控投喂封装（七段式/H3，权威在总控 short-drama-director/references/model-adapters.md）
  → ★ sd-qc-engine（连续性台账 → 剪辑转场分段门 → 声音设计 → 出稿机检）
  → 总控看板输出

按请求内容路由到对应工作流：

请求内容工作流规则库
跨镜/跨集一致性、站位/道具/服装核对连续性台账references/continuity-ledger.md
剪辑节奏、切点、转场、Seedance 15 秒分段剪辑转场分段门references/editing-transition-gate.md
对白/音效/BGM/环境声设计、台词时长复核声音设计references/sound-design-template.md
配 BGM、找音乐、AI 音乐提示词、版权检查后期音乐声音references/post-music-sound-design.md
出稿前机检、FAIL 清零、门禁口径查询出稿机检references/machine-check-manual.md
规则定级与冲突裁决口径验证治理references/rule-taxonomy-and-verdict.md
各阶段交付验收标准与校验方式阶段治理references/stage-acceptance-table.md

看片/拉片：先过 P0~P2 语义门禁（口径读总控 short-drama-director/references/quality-gate-review.md），再跑出稿机检兜底——顺序见 machine-check-manual.md §四。

四、工作流程（四种触发场景）

1. 连续性检查（跨镜/跨集）

  1. 读锁定剧本、资产锚点（按素材形态：M0 文字档案 / M1 编号册 + 接线单 / M2 资产图册）、分镜/投喂稿——任一必需输入缺失即停止并索要；但"没有图片"不属于"缺失"（本包不持有图，权威 = 总控 short-drama-director/references/asset-delivery-modes.md）。
  2. 逐场建立入口状态 → 变化 → 退出状态台账（人物账/场景账/时间轴账），按 continuity-ledger.md 模板。
  3. 核对每处状态变化有触发节拍、持续到明确解除；无「隐形换装/道具换手/姿态回退/站位漂移」。
  4. 与总控法则 20「人物账连续·摆位必入画」对账：摆了位的每个人本段镜行至少带到一次（C17/C17b/C17c/C17d 机检兜底——含跨段站位漂移 C17d）。
  5. 冲突只写报告并路由责任阶段，不代修。

2. 剪辑转场与分段检查

  1. 继承导演镜头策略（真一镜到底/伪一镜到底/常规切镜/混合），按 editing-transition-gate.md 铺连续时间线。
  2. Seedance 2.0 全能参考段 ≤15 秒硬门；2.5 段 ≤30 秒（C1 口径）。切点只落在许可窗口，禁切区零切镜。
  3. 长台词只在原有标点/换气/语义分句拆段，逐字顺序不变、不漏字不重字。
  4. 结构化 EDITING_PLAN 产物跑 scripts/validate_seedance_segment_map.py 校验（用法见规则库 §五）。

3. 声音设计

  1. 严格区分画内对白/画外音/旁白/电话音/系统音，逐字不改台词。
  2. 按 sound-design-template.md 分层（对白 > 动作音效 > 环境/BGM），对齐总控声音三层电平。
  3. 逐段复核 dialogue_slices：对白+换气+停顿+反应必须落在段长内；跨段台词同一 dialogue_id 无缝续接。
  4. Seedance 侧写法对齐总控 [SFX:] 标记与语气位规范（C5/C16）；H3 侧对齐 H10 声音分层。

4. 出稿机检

  1. 出稿前先过总控 P0~P2 门禁与 short-drama-director/references/prompt-feeding-checklist.md 挂线审查。
  2. 跑 python scripts/validate_prompt.py <稿> --model 2.5|2.0|h3（脚本在总控 scripts/）。
  3. FAIL 清零才交付；WARN 逐条修复或写明判定依据。完整操作见 machine-check-manual.md。

五、规则库索引

文件内容
references/continuity-ledger.md连续性台账模板：人物账（站位/朝向/服装/道具持拿）、场景账、时间轴账；与总控法则 20/C17 衔接
references/editing-transition-gate.md剪辑转场设计模板 + Seedance 15 秒分段门 + validate_seedance_segment_map.py 用法
references/sound-design-template.md声音设计模板：音效/人声/环境音/BGM 层级与写法；与总控 Seedance 原生音效标记体系衔接
references/post-music-sound-design.md后期音乐与声音设计：BGM Cue Sheet 标准格式 + 音乐搜索词生成规则（中英+素材库标签）+ AI 音乐生成提示词结构（Suno/Udio/ElevenLabs）+ 剪辑混音建议 + 静默点设计 + 版权与商用检查 + 声音体检清单。承接 sound-design-template.md 的 MUSIC 层 cue 落地执行
references/machine-check-manual.md出稿机检使用手册：C1C26 / H1H12 门禁说明、validate_prompt.py 命令行用法、P0~P2 与机检先后关系、check_package.py 包自检
references/rule-taxonomy-and-verdict.md规则三分类与统一裁决：硬约束/条件约束/创作建议的定义与判定三问、冲突裁决四级优先链、PASS/WARN/FAIL 三态跨阶段统一定义
references/stage-acceptance-table.md阶段交付验收表：6 阶段 × 交付物 × 校验方式三分类（机器可判 / AI 语义 / 人工看图看片）——防「声称完成」与「把语义判断题塞进正则」

六、共享契约声明

  • 锁系统（A/B/C/D 四锁）按总控 SKILL.md 〇节执行，本技能不重复定义。
  • 资产编号沿用总控「图N」体系与对应台账（按素材形态区分：M0《外观档案册》/ M1《资产编号册 + 素材接线单》/ M2《资产图册》；《剧组产出册》通用）。
  • 素材形态：本包不持有图。三档 M0/M1/M2（默认 M1），权威 = 总控 short-drama-director/references/asset-delivery-modes.md。本引擎的一致性门按形态改判——M0/M1 比文字档案（发/服装/体态/年龄感/标志物；场景坐标系与 SET_STATE），M2 才做图像级核验。禁止：索要图片作为继续条件 · 因缺图阻断交付 · 声称"已核验/与资产图册一致"而实际从未见过图（机检兜底见 sd-feed-engine/scripts/gate_check.py 闸门 6B）。
  • 投喂格式唯一权威 = 总控 short-drama-director/references/model-adapters.md（七段式）/ short-drama-director/references/h3-adapter.md（H3 六段式）。
  • Seedance 2.5 与 2.0 语法彻底分开禁混写；Seedance 侧时间轴整数秒口径（禁时间码、禁小数秒）；H3 侧切点按 H3 原生 MM:SS.mmm。
  • 改稿必跑总控「局部验证 → 联动冲突检查 → 全量审核」三步复验闭环并附《改动复验报告》。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Sd Qc Engine AI skill do?

【套装内部技能 · 全流程用户通常无需直接调用；直接说需求给总控 short-drama-director 即可，总控会自动路由到本技能】【漫剧老李 V7.4.26 套装 · 子技能】成片质检与连续性管理专精引擎：跨镜/跨集一致性台账、剪辑转场设计与 Seedance 15 秒分段门、声音设计模板、后期音乐与声音设计（BGM Cue Sheet/音乐搜索词/AI 音乐提示词/版权检查）、出稿机检使用。**一致性门按素材形态改判**（M0/M1 比文字档案；M2 才做图像级核验；本包不持有图，禁止索要图片、禁止虚假核验断言）。触发词：审查、质检、连续性检查、看片、拉片、台词诊断后复核、剪辑转场、分段检查、声音设计、机检、配BGM、后期音乐、找音乐、Suno提示词、音乐版权、素材形态。本技能为套装成员，完整全流程从总控 short-drama-director 进入。

Why use Sd Qc Engine on TypingMind?

Because you install it once and use it with any model. Sd Qc Engine is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Sd Qc Engine in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/lixiaoxiao9888-create/manju-laoli-skill/tree/main/sd-qc-engine. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Sd Qc Engine?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Sd Qc Engine?

As many as you like. As long as a model supports skills, you can use Sd Qc Engine with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Sd Qc Engine AI skill free?

Yes. It is published on GitHub by lixiaoxiao9888-create under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

Related AI Skills

View all
openclaw logo

Deslop

openclaw
OrganizationPopular

Diff-scoped AI-slop cleanup pass: strip comment slop, defensive-check slop, type-laundering, and style drift from the current branch diff before autoreview.

82.2K
391K
Instructions
View
openclaw logo

Discord Clawd

openclaw
OrganizationPopular

Use to talk to the Discord-backed OpenClaw agent/session; not for archive search.

82.2K
391K
1 files
View
openclaw logo

Discrawl

openclaw
OrganizationPopular

Discord archive: search, sync freshness, DMs, summaries, TUI, repo/release work.

82.2K
391K
1 files
View
openclaw logo

Gitcrawl

openclaw
OrganizationPopular

GitHub archive: issue/PR search, sync freshness, duplicate clusters, current GitHub handoff, and Gitcrawl repo work.

82.2K
391K
1 files
View

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇