Token导航 LogoToken导航TokenDH.com
效率操作浏览器clawhub未标认证来源可访问clear审计通过

map-route-reader地图路线阅读器

Agent Skill

map-route-reader 用于补充效率相关能力,适合在 OpenClaw 中需要让 Agent 承接效率相关任务时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

6,714

周安装

269

GitHub Stars

公开资料未说明

下载量

2,174
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:map-route-reader(地图路线阅读器)
来源仓库:https://github.com/wuyunting555/map-route-reader
安装命令:
openclaw skills install map-route-reader
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install map-route-reader

简介

直接读取高德 / 百度地图上的真实路线结果,返回用户能照着走的公交、地铁、打车或步行方案。适用于用户问怎么从 A 到 B、哪条线路更快、哪条路线更省事、打车是不是更方便、是否有更优替代路线时使用。重点不是给模糊估计,而是尽量给真实可执行路线。

SKILL.md

name
map-route-reader
description
|

Map Route Reader

产品定位

把“地图路线查询”变成更像结果型产品的 skill:不是只说个大概时间,而是尽量把真实路线读出来,整理成用户可以立刻参考的出行方案。

一句话卖点

直接读取高德 / 百度地图上的真实路线结果,优先返回用户能照着走的公交、地铁、打车或步行方案。

为什么更有下载欲望

很多同类路线能力看起来很强,但实际要:

  • 配 API
  • 补配置
  • 自己拼接流程
  • 最后还不一定能直接给出可走路线

这个 skill 主打的是:

  • 尽量直接可用
  • 优先读真实地图结果
  • 优先给可执行路线,不给空泛判断
  • 用户问哪条线更好时,会真的把两条路线拉出来做对比
  • 重点不是给模糊估计,而是给可直接拿去用的路线参考

你装它,想得到的是什么

  • 少一点“大概多久”的废话
  • 多一点“从哪上车、哪换乘、哪下车”的真实结果
  • 问打车 / 地铁 / 公交哪个更省事时,得到能直接参考的答案

Use this skill when the user asks for:

  • how to get from A to B
  • public transit routes
  • driving time
  • walking time
  • which station to transfer at
  • whether taxi is faster
  • whether a specific alternate route/line would be better
  • route comparisons such as “走 14 号线会不会更好”

Goal

Return a route the user can actually follow. Do not stop at vague estimates when a real route can be extracted.

Priority:

  1. exact route from a real map result
  2. if the user proposes an alternate route/line, extract that route too and compare side by side
  3. if exact route cannot be extracted, explain precisely where extraction failed and switch to the user-browser map tab via Browser Relay
  4. only as a temporary fallback, give a clearly-labeled estimate

Preferred workflow

0) For Amap, prefer uriapi route pages with lnglat when coordinates are available

If you can get origin/destination coordinates, prefer opening a direct Amap route URL like: https://ditu.amap.com/dir?type=bus&policy=2&from[lnglat]=<lng,lat>&from[name]=...&to[lnglat]=<lng,lat>&to[name]=...&src=uriapi&innersrc=uriapi&dateTime=now

Why:

  • this bypasses wrong default city context
  • the route card often renders directly in page text
  • total duration / walking distance / fare / line names can often be extracted from document.body.innerText

1) Try normal read first

  • Use browser on public map pages first.
  • Prefer Amap or Baidu Maps for China routes.
  • Try direct route pages with origin/destination query params.
  • If the page exposes the full route card, extract:

- total time - total walking time if visible - number of transfers - station names / line names - final arrival station or walking segment

1.5) On Amap route pages, extract hidden detail siblings

After the recommended route card renders, the full step-by-step transit detail may exist in the hidden sibling node right after the current .planTitle (for example an ol.p_route). Do not stop at total duration. Read that hidden detail block and extract:

  • walking segment distance and minutes
  • line name and direction
  • boarding station and exit/gate info
  • number of stops
  • transfer station
  • alighting station and exit number
  • final walking segment to destination
  • approximate duration of each transit segment when the page exposes enough information to infer it

2) If the map page needs interaction

If the site asks to confirm the start/end place, or the route card is hidden behind dynamic UI:

  • keep using browser
  • interact to confirm the correct start/end point
  • prefer the first recommended public transit route unless the user asked for fastest/least walking/least transfers

3) For China map sites, prefer the user's real map tab when city context matters

For Amap/Baidu route pages, the isolated browser may resolve the wrong city or fail to expand the route card. If the page shows the wrong city, cannot resolve start/end places, or keeps asking to choose the correct place, switch early to the user's map tab via Browser Relay and read the live route result there.

4) If normal browser still cannot read the real route

Switch to the user’s browser tab via Browser Relay.

  • Ask the user to open the target map route page in Chrome.
  • Ask them to click Browser Relay on the exact map tab.
  • Then read the real route card from that tab.

Comparison rule

If the user asks whether a specific route/line would be better, for example:

  • “走 14 号线会不会更好”
  • “打车会不会更方便”
  • “xx 这么走是不是更好”

then you must:

  1. keep the current recommended route
  2. extract the user-specified route too
  3. present a side-by-side comparison

Do not guess. Do not answer from intuition alone. Do not say “应该不会” or “大概率更慢” unless the alternate route has been pulled out and compared.

Output format

For a single route, return in this structure:

  • 起点:...
  • 终点:...
  • 方式:公交 / 地铁 / 打车 / 步行
  • 预计总时长:...
  • 推荐路线:...
  • 换乘:...
  • 每段步骤:...
  • 下车后:...
  • 备注:如有不确定,明确写出来

If route detail is available, include:

  • 上车站
  • 下车站
  • 换乘站
  • 出口
  • 每段步行距离与时间
  • 每段乘车站数
  • 每段乘车大约多久

For a route comparison, return side-by-side:

  • 方案 A
  • 方案 B
  • 总时长
  • 步行距离
  • 换乘次数
  • 线路结构
  • 哪个更快 / 哪个更省事 / 哪个更少走路

Rules

  • Do not stop at “I’ll check” or “probably around ...” unless exact extraction has genuinely failed.
  • If exact extraction failed, say which step failed.
  • For map queries, a user-visible answer should try to be route-usable.
  • If the user corrected the place name, restart from the corrected place immediately.
  • If the user corrected the place name, never reuse the old route result as if it applied to the corrected place.
  • For comparison questions, do not brain-fill the alternate route. Pull it out, then compare.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

88.6%
按下载量换算1,926

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills