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审计提醒

lightcone-session光锥会话

Agent Skill

lightcone-session 用于处理浏览器自动化、网页检查和页面信息提取,适合在 OpenClaw 中需要让 Agent 打开页面、读取网页或验证前端流程时使用。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

7,075

周安装

289

GitHub Stars

公开资料未说明

下载量

2,289
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:lightcone-session(光锥会话)
来源仓库:https://github.com/eddieogola/lightcone-session
安装命令:
openclaw skills install lightcone-session
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 lightcone-session

简介

细粒度浏览器控制会话,支持手动逐步操作流程引导。

  • 适合登录认证、多步骤表单提交与交互式测试任务。
  • 提供屏幕截图与 DOM 快照便于问题排查与结果验证。
  • 会话状态保存在本地,关闭后需重新初始化连接。lightcone-session 属于效率类 Skill,可作为该场景下的辅助能力补充。
  • 涉及敏感操作时请确保环境安全,避免凭证泄露风险。

SKILL.md

name
lightcone-session
description
|
user-invocable
true
metadata
{"openclaw":{"emoji":"🖥️","homepage":"https://docs.lightcone.ai","requires":{"env":["TZAFON_API_KEY"]},"primaryEnv":"TZAFON_API_KEY"}}

Lightcone Session Control

Three tools for manual step-by-step control of a cloud computer. Use when you need precise multi-step interaction with a website or desktop application.

For simple "go do this" tasks, prefer lightcone-browse (the /lightcone-browse command) instead.

Tools

ToolPurpose
lightcone_session_createCreate a cloud computer (browser or desktop)
lightcone_session_actionSend actions (click, type, scroll, screenshot, etc.)
lightcone_session_closeShut down the computer

Workflow Pattern

1. lightcone_session_create  → get session ID + initial screenshot
2. lightcone_session_action  → navigate, click, type, screenshot (repeat)
3. lightcone_session_close   → shut down

lightcone_session_create

ParameterRequiredDescription
urlNoURL to navigate to immediately
kindNo"browser" (default) or "desktop"

Returns: session ID (e.g., lc-a1b2c3d4) + screenshot of initial state.

lightcone_session_action

ParameterRequiredDescription
sessionIdYesSession ID from create
actionYesOne of the actions below
urlFor navigateTarget URL
x, yFor click/scrollPixel coordinates
textFor typeText to enter
keysFor hotkeyComma-separated keys, e.g. "Control,c"
dx, dyFor scrollScroll deltas
secondsFor waitDuration in seconds
commandFor debugShell command to run

Actions

ActionReturnsUse for
screenshotScreenshotSee current state
navigateScreenshotGo to a URL
clickScreenshotClick at (x, y)
doubleClickScreenshotDouble-click at (x, y)
rightClickScreenshotRight-click for context menu
typeScreenshotEnter text at cursor
hotkeyScreenshotKeyboard shortcut (e.g., Enter, Tab, Control+a)
scrollScreenshotScroll at position
htmlPage HTMLExtract page source
waitScreenshotPause before next action
debugCommand outputRun shell command in the computer

Example: Login and extract data

Step 1: Create a cloud computer at the login page
  lightcone_session_create { url: "https://app.example.com/login" }

Step 2: Type username (click the email field first)
  lightcone_session_action { sessionId: "lc-...", action: "click", x: 640, y: 300 }
  lightcone_session_action { sessionId: "lc-...", action: "type", text: "user@example.com" }

Step 3: Type password
  lightcone_session_action { sessionId: "lc-...", action: "hotkey", keys: "Tab" }
  lightcone_session_action { sessionId: "lc-...", action: "type", text: "password123" }

Step 4: Submit
  lightcone_session_action { sessionId: "lc-...", action: "hotkey", keys: "Return" }
  lightcone_session_action { sessionId: "lc-...", action: "wait", seconds: 3 }

Step 5: Screenshot to verify login succeeded
  lightcone_session_action { sessionId: "lc-...", action: "screenshot" }

Step 6: Extract page content
  lightcone_session_action { sessionId: "lc-...", action: "html" }

Step 7: Shut down
  lightcone_session_close { sessionId: "lc-..." }

Tips

  • Always take a screenshot after navigation to see the current state before clicking
  • Use coordinates from the screenshot to target clicks accurately
  • Wait 2-3 seconds after navigation or form submission for pages to load
  • Use html action to extract structured data from the page
  • Always close sessions when done to free resources

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

81.71%
按下载量换算1,870

安全审计

VirusTotal

可疑

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。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