Token导航 LogoToken导航TokenDH.com
研究检索需要联网clawhub未标认证来源可访问clear审计通过

robotaxi-tracker机器人出租车追踪器

Agent Skill

robotaxi-tracker 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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

2,763

周安装

114

GitHub Stars

公开资料未说明

下载量

903
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:robotaxi-tracker(机器人出租车追踪器)
来源仓库:https://github.com/pekky/robotaxi-tracker
安装命令:
openclaw skills install robotaxi-tracker
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 robotaxi-tracker

简介

当用户询问最新或当前的美国自动驾驶汽车、自动驾驶汽车、机器人出租车、特斯拉或 Waymo 车辆数量、车队规模、城市信息时,请使用此技能。

SKILL.md

name
robotaxi-tracker
description
Use this skill when the user asks for the latest or current US autonomous vehicle, self-driving car, robotaxi, Tesla, or Waymo vehicle counts, fleet size, city distribution, official fleet numbers, tracker-collected counts, or comparison tables, especially when the data should come from robotaxitracker.com / Robotaxi Tracker. It covers natural requests like "帮我查下最新的美国自动驾驶车辆数量" and provides the workflow to inspect frontend bundles, discover Convex backend queries, separate official-fleet vs tracker-collected metrics, validate city counts and totals, and format compact comparison tables.

Robotaxi Tracker

Use this skill for requests about robotaxitracker.com, especially when the user wants:

  • latest US autonomous vehicle counts
  • current self-driving car / robotaxi counts in the US
  • Tesla vs Waymo fleet comparison
  • city distribution
  • official fleet numbers
  • tracker-collected / discovered vehicle counts
  • output in a compact table

This skill should also trigger for natural-language requests like:

  • 帮我查下最新的美国自动驾驶车辆数量
  • 帮我查下最新的美国 robotaxi 数量
  • 最新的 Tesla 和 Waymo 车队数量
  • 美国自动驾驶车辆城市分布

Core rules

  1. Do not rely on HTML alone. Inspect JS bundles and query the public backend.
  2. Do not mix metric types. Separate:

- official fleet - tracker-collected / discovered vehicles

  1. Validate:

- service area count - row count in the city table - city sum vs chosen aggregate metric

  1. If the user requests a final table only, output only the table.
  2. Do not assume the first Convex host you find is valid. Test candidate hosts and keep only hosts that return status: success for the expected public queries.
  3. Do not claim the backend requires authentication unless you have confirmed a direct API response that explicitly indicates auth is required. A 404 or wrong host is not an auth failure.

Required tools

  • curl
  • rg
  • jq

If network is restricted, switch to a tool or environment that can fetch the site.

Workflow

1. Fetch homepage HTML

curl -sS https://robotaxitracker.com/ -o /tmp/robotaxi_home.html

2. Extract and download homepage JS bundles

You must scan all homepage chunks. Do not pick a single chunk by guesswork.

grep -o '/_next/static/chunks/[^"?]*\.js' /tmp/robotaxi_home.html | sort -u > /tmp/robotaxi_chunks.txt
cat /tmp/robotaxi_chunks.txt
mkdir -p /tmp/robotaxi_js
while read -r u; do
  [ -n "$u" ] || continue
  f=$(basename "$u")
  curl -sS "https://robotaxitracker.com$u" -o "/tmp/robotaxi_js/$f"
done < /tmp/robotaxi_chunks.txt

If you have not scanned all chunks listed in /tmp/robotaxi_chunks.txt, stop and finish that first.

3. Find backend query names and enumerate all candidate hosts

rg -o 'api\.[A-Za-z0-9_.]+' -n /tmp/robotaxi_js/*.js | sort -u
rg -o '[A-Za-z0-9-]+\.convex\.cloud' -n /tmp/robotaxi_js/*.js | sort -u > /tmp/robotaxi_hosts.txt
cat /tmp/robotaxi_hosts.txt
rg -n 'convex\.cloud|ConvexReactClient' /tmp/robotaxi_js/*.js

Look for queries like:

  • api.queries.serviceAreas.list
  • api.queries.fleet.getHomepageData
  • api.queries.fleet.getRecentlyAddedCounts

Known current host notes:

  • Prefer graceful-eel-151.convex.cloud first. It was verified on 2026-04-10 to return status: success for the public queries above.
  • happy-otter-123.convex.cloud was observed returning 404 Not Found on 2026-04-10. Treat it as stale or invalid unless re-verified.

Do not stop after finding one host. Enumerate all candidate hosts first.

4. Validate all candidate hosts before choosing one

Before using any discovered Convex host for real data extraction, probe every candidate host with a known public query.

while read -r host; do
  [ -n "$host" ] || continue
  echo "=== $host ==="
  curl -sS "https://$host/api/query"     -H 'Content-Type: application/json'     --data '{"path":"queries/serviceAreas:list","args":[{"provider":"waymo"}]}'
  echo
done < /tmp/robotaxi_hosts.txt

Selection rules:

  1. You must test every host in /tmp/robotaxi_hosts.txt before choosing one, unless the first verified-success host is graceful-eel-151.convex.cloud.
  2. Choose the first host that clearly returns {"status":"success", ...} for the expected public query.
  3. If a host returns 404 Not Found, empty output, HTML, or malformed JSON, treat it as invalid and continue to the next candidate host.
  4. Do not describe 404, empty output, or wrong-host behavior as an authentication requirement.
  5. Only describe the backend as auth-protected if the API response explicitly indicates authentication or authorization failure.
  6. If graceful-eel-151.convex.cloud returns status: success, prefer it immediately.

After choosing a host, keep using that same validated host for all subsequent queries in the run.

5. Get service areas

curl -sS https://<convex-host>/api/query \
  -H 'Content-Type: application/json' \
  --data '{"path":"queries/serviceAreas:list","args":[{"provider":"waymo"}]}'

Extract rows:

curl -sS https://<convex-host>/api/query \
  -H 'Content-Type: application/json' \
  --data '{"path":"queries/serviceAreas:list","args":[{"provider":"waymo"}]}' \
  | jq -r '.value[] | [.name,.slug,.id] | @tsv'

Repeat for tesla.

6. Tracker-collected counts

Homepage aggregate:

curl -sS https://<convex-host>/api/query \
  -H 'Content-Type: application/json' \
  --data '{"path":"queries/fleet:getHomepageData","args":[{"provider":"waymo"}]}'

Per city:

curl -sS https://<convex-host>/api/query \
  -H 'Content-Type: application/json' \
  --data '{"path":"queries/fleet:getHomepageData","args":[{"provider":"waymo","serviceAreaId":"<serviceAreaId>"}]}'

Use .value.totalVehiclesCount for the per-city tracker count unless the user explicitly asks for another field.

7. Official fleet counts

Search bundles:

rg -n 'getOfficialFleetCount|getTotalOfficialFleet|officialFleet|official fleet' /tmp/robotaxi_js/*.js

Important:

  • the site may expose official fleet only for some providers
  • official total and official city mapping may not cover exactly the same set of cities
  • do not infer missing city numbers unless the user explicitly asks for estimation

8. Validation

Always check:

  1. number of service areas
  2. number of city rows in the final table
  3. sum of chosen city metric
  4. chosen total metric
  5. the selected Convex host and whether it returned status: success during validation

If totals differ, mention the mismatch only if the user asked for explanation. Otherwise, use the requested metric and keep output compact.

If you cannot produce live data, explain exactly which host was tested, what raw failure happened, and why that failure implies the next step. Never jump directly from a wrong-host response to an authentication conclusion.

Output defaults

If the user asks for a combined comparison table, use:

城市Waymo数量Tesla数量
Bay Area......
Austin......
总数量......

Rules:

  1. Put the same city on the same row.
  2. Fill missing values with 0.
  3. Add 总数量 as the last row.
  4. Do not include extra prose if the user asked for table-only output.

Common mixed-metric case

If the user says:

  • Waymo uses official fleet
  • Tesla uses site/tracker data

Then:

  1. fetch Waymo official city numbers from frontend-exposed config
  2. fetch Tesla tracker city numbers from backend query
  3. merge by city
  4. output one table only

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

86.92%
按下载量换算785

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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