Summary
规划、编写、审阅和修复由真实授权素材与 AI 生成镜头混合制作的电影级文旅宣传片。用于文旅短片创意简报、场景图筛选、首帧空间分析、分镜脚本、画面内容提示词、人物动作、导演级运镜、自然动机布光、转场与尾帧衔接、Seedance 2.0/可灵/Midjourney…
kangarooking/director-skills · Archived
规划、编写、审?
npx skills add kangarooking/director-skills --skill travel-skill
规划、编写、审阅和修复由真实授权素材与 AI 生成镜头混合制作的电影级文旅宣传片。用于文旅短片创意简报、场景图筛选、首帧空间分析、分镜脚本、画面内容提示词、人物动作、导演级运镜、自然动机布光、转场与尾帧衔接、Seedance 2.0/可灵/Midjourney…
This repository is archived — consider an actively maintained alternative.
Deep travel research workflow for museums and ancient architecture. Input a city name, auto-gen…
6.1K installsThis skill should be used whenever users need help planning trips, creating travel itineraries,…
3.8K installsTravel destination research and daily itinerary creation with logistics planning, budget tracki…
488 installsUse when designing animations for travel booking, hospitality apps, tourism platforms, or vacat…
464 installsRelated neighbors and high-traction skills in the same topics — useful to compare before installing.
Transform any AI agent into a world-class travel planner using Kontour AI's 9-dimension progres…
463 installsCreates comprehensive travel itineraries with destination analysis, budget planning, transporta…
369 installsTravel planning with real place data, interactive maps, and persistent shareable itineraries at…
301 installsPlan trips, search flights/hotels on Expedia, discover destinations via TikTok, Instagram, and …
285 installs외교부 0404 국가·지역별 여행경보 Open API를 k-skill-proxy 경유로 조회해 국가별 공식 경보 단계와…
38 installs分析?
113 installsDeclared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.
Alternate registries and mirrors of this skill.
npx skills add tencentcloudbase/awesome-miniprogram-skills --skill travel-skill
main
Files included with this skill beyond the listing page.
SKILL.md
6,220 B
SUMMARY.md
505 B
把提示词当作可执行的摄影调度,不把它写成镜头愿望清单。先确认故事、素材和首帧所允许的空间,再设计画面内容、人物动作、摄影机和光线。
记录目标、受众、平台、总时长、画幅、叙事主线、情绪、真实素材、参考人物、图片模型、视频模型及单次时长。Seedance 2.0 和可灵均可使用首尾帧;默认只在确有落点控制需要时使用尾帧。
完整项目读取 [workflow-and-gates.md](references/workflow-and-gates.md),并复制 [project-brief.md](assets/templates/project-brief.md)。
优先让真实授权素材承担地貌、雪山、湖泊、峡谷、村落、动物和人文纪实;让 AI 承担固定角色剧情、难以补拍的动作和受控转场。记录水印、分辨率、黑边、来源、人物与文化风险。
在写任何画面内容提示词或运镜前,完整读取 [source-frame-spatial-audit.md](references/source-frame-spatial-audit.md),填写 [shot-manifest.yaml](assets/templates/shot-manifest.yaml) 中的空间字段。
必须回答:
缺少关键首帧时,先列出待补场景图,不用文字强迫模型创建未知空间。
按“叙事目标 → 分时段画面内容 → 景别与构图 → 人物动作 → 摄影机调度 → 光影 → 声音/台词 → 尾帧”的顺序写。画面内容必须分时段,但各时段描述同一连续空间中的状态变化,不能写成多个剪辑镜头。
运镜前完整读取 [camera-motion-library.md](references/camera-motion-library.md);布光前完整读取 [motivated-lighting-library.md](references/motivated-lighting-library.md);涉及跨镜衔接时读取 [transitions-and-continuity.md](references/transitions-and-continuity.md)。
完整读取 [model-recipes.md](references/model-recipes.md)。中文描述中保留准确英文术语。Midjourney 提示词控制在用户或模型给定字符数内;Seedance 2.0 与可灵的单段时长以当前任务参数为准,不假定永远为 15 秒。
采用三轮迭代:
出现乱镜头时,先删掉第二条轨迹和无关风格词,不继续堆形容词。
用 [failure-catalog.md](references/failure-catalog.md) 对生成结果做空间、动作、运镜、人物、光线、连续性、技术和风险八类检查。运行:
python3 scripts/lint_prompt.py prompt.txt --model seedance --max-chars 0
python3 scripts/validate_timeline.py storyboard.md --max-shot-seconds 15
python3 scripts/lint_shot_manifest.py shot-manifest.yaml
修复局部镜头时保持镜号稳定;只更新时间码和明确受影响的依赖。
完整制片默认在以下节点暂停:
用户明确要求批量执行时可以减少确认,但不得跳过首帧空间审计。
单镜修改时只输出该镜;完整项目优先输出:
00-project-brief.md01-asset-register.csv02-visual-reference.md03-storyboard.md04-shot-manifest.yaml05-generation-prompts.md06-generation-log.csv07-qc-report.md西藏项目还必须读取 [tibet-reference.md](references/tibet-reference.md)。