← Back to Skills Marketplace
ccshen722

light-chaser

by ccshen · GitHub ↗ · v1.0.2 · MIT-0
cross-platform ⚠ suspicious
90
Downloads
2
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install light-chaser
Description
旅行摄影动态脚本助手,解决旅行者"去哪拍、何时拍、怎么拍、怎么走"四大核心痛点。将碎片化攻略转化为可执行的拍摄时间线。当用户提到旅行拍照、打卡、旅拍、摄影攻略、出行拍摄计划时使用此技能。
README (SKILL.md)

Light Chaser · 旅行光影追逐者

工作流程

第一步:收集出行信息

必须向用户收集以下信息(用友好的对话方式,非清单式询问):

必填:

  • 🗺️ 目的地:城市/景区名称
  • 📅 出行时间:具体日期(用于天气查询)
  • 📷 拍摄设备:手机/相机/无人机等(可多选)

选填:

  • 🎨 喜欢的风格:日系/港风/ins风/胶片/暗黑/纯欲/国风等(若用户未提及则根据设备和目的地智能推断)

第二步:搜索景点拍摄攻略

收集用户信息后,立即使用搜索引擎检索目的地的真实打卡推荐数据,作为景点选取的核心依据:

搜索关键词(按优先级依次搜索):

  1. {目的地} 旅游打卡点拍照推荐
  2. {目的地} 旅拍景点 ins风 出片
  3. {目的地} 摄影攻略 必去地点

从搜索结果中提取以下信息:

  • 高频出现的热门打卡地(出现 3 次以上优先纳入)
  • 各景点的最佳拍摄时段(若有提及)
  • 景点间的地理位置关系(用于规划最优路径)
  • 近期是否有季节性特色(花期/雪景/灯会等)

搜索失败处理: 若搜索无结果或网络异常,回退使用已知的目的地经典景点知识继续生成。


第三步:查询天气信息

收集信息后,读取技能目录下的 config.env 决定查询方式:

CONFIG="「技能目录」/config.env"  # 即 SKILL.md 所在目录
QWEATHER_HOST=""
QWEATHER_KEY=""
if [ -f "$CONFIG" ]; then
  while IFS='=' read -r key value; do
    case "$key" in
      QWEATHER_HOST) QWEATHER_HOST="$value" ;;
      QWEATHER_KEY)  QWEATHER_KEY="$value"  ;;
    esac
  done \x3C "$CONFIG"
fi

方式 A:环境变量存在(QWEATHER_HOSTQWEATHER_KEY 均非空)→ 调用 API

1. 城市 ID 查询

curl --compressed "https://${QWEATHER_HOST}/geo/v2/city/lookup?location={城市名}&key=${QWEATHER_KEY}&lang=zh"

取返回值中第一个结果的 id 字段作为 LOCATION_ID

2. 逐天天气预报(最多7天)

curl --compressed "https://${QWEATHER_HOST}/v7/weather/7d?location={LOCATION_ID}&key=${QWEATHER_KEY}&lang=zh"

3. 逐小时天气预报

curl --compressed "https://${QWEATHER_HOST}/v7/weather/24h?location={LOCATION_ID}&key=${QWEATHER_KEY}&lang=zh"

4. 日出日落时间

curl --compressed "https://${QWEATHER_HOST}/v7/astronomy/sun?location={LOCATION_ID}&date={YYYYMMDD}&key=${QWEATHER_KEY}&lang=zh"

方式 B:环境变量不存在 → 通过网络搜索获取天气

使用 fetch_web / 搜索引擎搜索以下信息:

  • 搜索词:{目的地} {出行日期} 天气预报
  • 目标获取:温度范围、天气现象、降水概率、风力、湿度
  • 日出日落:搜索 {目的地} {出行日期} 日出日落时间 或根据天文规律估算

注意:方式 B 获取的数据精度低于 API,在天气概况中注明「数据来源:网络搜索,仅供参考」

天气解读规则:

综合 textDay(天气现象)+ cloud(云量 0–100)+ vis(能见度)+ uvIndex(紫外线)进行综合评级:

天气状况 云量参考 拍摄评级 光线描述与拍摄策略
晴天 cloud \x3C 20 ⭐⭐⭐⭐⭐ 黄金时段(日出后1h/日落前1h)光线最佳;正午 uvIndex≥6 时转室内或有遮阴处拍摄
晴间多云 cloud 20–50 ⭐⭐⭐⭐⭐ 云块遮住太阳时形成天然柔光,追云间隙出现的光束,黄金时段层次丰富
多云 cloud 50–80 ⭐⭐⭐⭐ 全天候柔光箱,色彩还原准确;全天皆可拍,无需刻意避开正午
阴天/薄雾 cloud > 80 ⭐⭐⭐ 高级滤镜效果,适合氛围感/日系风;vis 低时前景感和层次感更突出
小雨 ⭐⭐⭐ 雨后 30 分钟路面反光绝佳;可借助雨伞/雨衣造型;注意设备防水
大雨/暴雨 建议将户外景点调整为室内;提醒设备防水保护
雪天 ⭐⭐⭐⭐⭐ 稀缺场景,优先户外雪景;曝光补偿 +1EV 防止雪面欠曝

第四步:生成拍摄时间线

结合搜索到的景点数据 + 天气 + 日出日落时间,输出完整拍摄脚本。景点选取优先采用搜索结果中的高频推荐地,而非固定模板。

时间线格式模板:

## 📅 {日期} · {目的地} 拍摄脚本
### 🌤️ 今日天气概况
{天气摘要:温度范围、天气现象、特殊提示}

---

### ⏰ 拍摄时间线

#### 🌅 时间点:{开始时间} - {结束时间}
📍 **地点**:{具体地点名称}
⭐ **推荐星级**:{★★★★★ / ★★★★☆ / ★★★☆☆ 等,结合当日天气+光线条件+景点出片潜力综合评定,并用一句话说明理由,如:★★★★★ 黄金时段逆光+湖面晨雾,全天最佳机位}
⏱️ **建议停留**:{X 分钟 / X 小时,说明理由,如:光线窗口仅45分钟,建议不超时}
🌞 **光线状态**:{光线描述,如:侧逆光/黄金光/漫反射柔光;结合云量说明,如:云量60%,柔光均匀}
📷 **拍摄技巧**:
  - {构图方法,如:利用XXX做前景,长焦压缩背景}
  - **设备参数建议**:
    - 手机:{ISO/曝光补偿/模式建议}
    - 相机:{推荐焦段,如:24-35mm广角适合大场景/85-135mm人像压缩/70-200mm长焦压缩背景} + {光圈/快门/ISO参考值}
    - 无人机:{飞行高度/角度/注意事项}(仅当用户有无人机时)
👗 **穿搭建议**:{推荐色系及理由,如:推荐白色/红色系,与灰色建筑形成对比}

---

时间节点规划原则:

  • 黄金时段(日出后60分钟内,日落前60分钟):必须安排重点户外景点
  • 蓝调时刻(日落后15-30分钟):安排城市夜景/有灯光的地标
  • 正午(11:00-14:00):安排室内景点/有遮挡的场所/餐厅打卡
  • 按"体力消耗最优路径"排序景点,减少无效往返

第五步:后期修图建议

在时间线末尾附上个性化后期建议:

## 🎨 后期修图方向

### 整体基调
{根据天气+风格推荐:如 "今日多云柔光,适合日系小清新调色"}

### 参数方向
- **亮部**:{建议,如:适当压低高光,防止过曝}
- **阴影**:{建议,如:提亮阴影+0.3,保留细节}
- **色温**:{建议,如:偏暖+200,增加黄金时段氛围}
- **色调**:{建议,如:绿色饱和度-20,青色色相偏蓝,打造日系感}
- **颗粒**:{按风格决定,如:添加轻微颗粒感,增强胶片质感}

### 调色关键词
{给出 2–3 个描述性调色方向关键词,不提具体软件名和预设名,如:「冷暖对比拉开」「蓝调压暗部」「胶片褪色感」,用户可自行在任意修图工具中实现}

### 注意避坑
- {1-2条常见问题,如:多云天sky replacement慎用,否则违和感强}

关键注意事项

  1. 无人机提醒:若用户有无人机,必须提示查询目的地禁飞区(使用大疆飞行地图 https://www.dji.com/cn/flysafe
  2. 时区处理:API 返回时间为北京时间(UTC+8),直接使用
  3. 景点数量:根据出行天数合理安排,单天不超过4-5个景点
  4. 设备针对性:用户没有某类设备时,不要输出该设备的参数建议
  5. 风格一致性:穿搭建议、后期建议三者风格要统一
Usage Guidance
This skill looks coherent and focused on producing travel photography timelines. Before installing: (1) confirm you are comfortable the agent will run web searches and (optionally) make HTTP calls to a weather API; (2) only add a QWEATHER_KEY to config.env if you obtained it from a trusted account and you accept storing that key in the skill directory; (3) ensure your runtime has network access (and curl if you expect the shell snippets to run), and understand that destination/date queries will be sent to search providers and the optional weather host; (4) if you need offline or privacy-preserving use, do not populate config.env and avoid invoking the network-dependent features.
Capability Analysis
Type: OpenClaw Skill Name: light-chaser Version: 1.0.2 The skill functions as a travel photography assistant but employs high-risk capabilities including direct shell-like logic and network access via `curl` to the QWeather API. It reads local configuration data from `config.env` and constructs shell commands using external variables, which presents a potential surface for command injection or unauthorized network activity. While these actions are aligned with the stated purpose, the use of shell execution and network requests meets the threshold for a suspicious classification under the provided criteria.
Capability Assessment
Purpose & Capability
Name/description (travel photography timeline) align with the instructions: collecting user trip info, web searching for popular photo spots, querying weather, and producing a shooting timeline. Requested/optional artifacts (config.env for a weather API) are appropriate for the stated purpose.
Instruction Scope
SKILL.md explicitly directs the agent to (a) perform web searches for destination-specific recommendations, (b) optionally call a weather API if QWEATHER_HOST/KEY are provided, and (c) read the skill's local config.env file. These actions are within scope, but the skill instructs using curl and shell snippets even though no required-binaries list is declared—so the runtime environment must permit HTTP requests and optionally curl execution. Also be aware search queries will include destination/date info (potentially sensitive).
Install Mechanism
Instruction-only skill with no install steps and no code files—nothing will be written to disk beyond user-edited config.env. This is the lowest-risk install model.
Credentials
No required environment variables or credentials are declared. The only optional credential is a QWeather host/key read from the local config.env—this is proportional to the stated goal of getting precise weather data. Users should treat QWEATHER_KEY as a secret and only populate it if they trust the API provider.
Persistence & Privilege
The skill is not always-enabled, does not request persistent system privileges, and does not modify other skills or global agent settings. It only reads a local config.env if present.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install light-chaser
  3. After installation, invoke the skill by name or use /light-chaser
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
light-chaser 1.0.2 - 新增 config.env 配置文件,支持 API host 和 key 信息本地化管理 - 天气信息查询逻辑调整:优先读取 config.env 配置,无则自动降级为网络搜索 - 文件读取方式优化,减少环境变量遗漏导致的查询失败 - 搜索景点推荐关键词更通用,去除固定年份限制 - SKILL.md 文档对天气查询与优先级说明更加清晰
v1.0.1
rename
v1.0.0
light-chaser 1.0.0 初始发布 - 全新旅行拍摄时间线脚本生成器,自动整合路线、最佳拍摄时段、设备建议与调色方向 - 智能收集出行信息:目的地、日期、拍摄设备、风格偏好 - 实时搜索目的地最新打卡景点与拍摄攻略,优先高频推荐地 - 接入天气和日出日落API,生成当天光线与天气分析,调优每个景点的拍摄建议 - 输出详细拍摄时间线、穿搭建议、设备设置、无人机提醒及后期修图方案 - 针对不同摄影设备个性化参数推荐,风格建议全流程一致
Metadata
Slug light-chaser
Version 1.0.2
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is light-chaser?

旅行摄影动态脚本助手,解决旅行者"去哪拍、何时拍、怎么拍、怎么走"四大核心痛点。将碎片化攻略转化为可执行的拍摄时间线。当用户提到旅行拍照、打卡、旅拍、摄影攻略、出行拍摄计划时使用此技能。 It is an AI Agent Skill for Claude Code / OpenClaw, with 90 downloads so far.

How do I install light-chaser?

Run "/install light-chaser" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is light-chaser free?

Yes, light-chaser is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does light-chaser support?

light-chaser is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created light-chaser?

It is built and maintained by ccshen (@ccshen722); the current version is v1.0.2.

💬 Comments