Token导航 LogoToken导航TokenDH.com
研究检索敏感数据clawhub未标认证来源可访问clear审计提醒

voxrnvoxrn 搜索

Agent Skill

voxrn 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

1,248

周安装

51

GitHub Stars

公开资料未说明

下载量

404
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:voxrn(voxrn 搜索)
来源仓库:https://github.com/robit-man/voxrn
安装命令:
openclaw skills install voxrn
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 voxrn

简介

打通多种即时通讯渠道的电话拨打、短信发送与联系人检索服务桥接器。

  • 兼容Slack、Telegram、WhatsApp等平台的消息路由与调度任务执行功能。
  • 实现跨应用通信协议的统一入口管理降低多端维护复杂度负担。
  • 使用前需配置各渠道Webhook地址并获取相应API访问权限凭证。
  • 集成进OpenClaw后可统一调用send_message等接口实现多渠道触达目标用户群。

SKILL.md

name
voxrn
description
Place phone calls, send SMS, search contacts, and run agent dispatches via Voxrn. Bridges any OpenClaw-bridged chat (Slack, Telegram, iMessage, WhatsApp, Discord, more) to a real PSTN.
metadata
{"clawdbot":{"emoji":"📞","requires":{"bins":["openclaw"]},"links":{"home":"https://voxrn.com","docs":"https://voxrn.com/docs/integrations/openclaw"}}}

Voxrn 📞

Voxrn is a browser-native voice + SMS platform with an MCP-driven agent runtime. This skill registers Voxrn as an outbound MCP server in your OpenClaw workspace, so the agent can place calls, send messages, manage contacts, and watch live agent sessions from any bridged chat.

When to use this skill

Reach for these tools when the user asks to:

  • place a phone call — "call my plumber", "ring the front desk", "dial +1 415 555 0188"
  • send a text message — "text Mina that I'm running 5 min late"
  • find a contact — "what's the number for Atlas Freight?"
  • save someone to contacts — "save +1 415 555 0188 as Front Desk"
  • see what's on the wire — "are any agent calls running?"
  • end a runaway agent call — "kill the call to Mina"
  • check the balance — "what's my Voxrn balance"

Tool catalog

ToolArgsUse it for
call.place{ to: E.164, taskBrief?: string, voiceProfileId?: string }Dial out — agent runs the call following the task brief
call.list_activenoneShow currently-live calls in the workspace
call.end{ callSid: string }Hard-kill an active call by SID
transcript.stream{ callSid: string }Subscribe to live captions for a call
message.send{ to: E.164, body: string }Send an SMS
contact.search{ q: string }Free-text search across name, number, tag
contact.upsert{ name, number, tags?, notes? }Create or update a contact
balance.checknoneRead the workspace balance in USD

Configuration

Set these env vars (or use OpenClaw's secrets manager):

export VOXRN_API_KEY="vxk_..."   # mint at https://voxrn.com/enterprise/dashboard/agents
export VOXRN_BASE_URL="https://voxrn.com"  # optional; defaults to prod

Or:

openclaw config secrets set voxrn_api_key vxk_...

The skill's mcp.json reads ${VOXRN_API_KEY} at MCP-registration time so the secret never reaches the agent's prompt window.

Conventions the agent should follow

  • Confirm the destination first. If the user says "call Atlas", run contact.search { q: "Atlas" } and ask which match before placing the call.
  • Quote the cost before running call.place. Default rate-card multiplier is 1.25× over carrier; international calls can be more. Either ask "ok if this costs ~$X for a 2-minute call?" or run balance.check first.
  • Pass the user's prose into taskBrief. Don't reformat — the brief travels straight to the agent runtime.
  • Confirm before call.end unless the user explicitly said "kill it now". Action is destructive and audit-logged.
  • Never echo the API key. The skill registers the server with ${VOXRN_API_KEY} so the agent never sees the secret.

Examples

Place a call

*User:* call my plumber and ask if Thursday morning works for the kitchen sink
agent: contact.search { q: "plumber" }
→ [{ id: "c-12", name: "Atlas Plumbing", number: "+14155551234", tags: ["plumber"] }]

agent: balance.check {}
→ { usd: 23.40 }

agent: "Calling Atlas Plumbing at +1 415 555 1234 to ask about Thursday morning availability. Estimated cost: ~$0.04. Going ahead."

agent: call.place {
  to: "+14155551234",
  taskBrief: "Call Atlas Plumbing on the user's behalf. Ask if Thursday morning is available to fix a kitchen sink leak. If yes, confirm a 9am window. If no, ask what's their next opening this week."
}
→ { callSid: "CAabc...", reservedUsd: 0.04 }

agent: "Call placed. SID CAabc... — I'll let you know when it ends."

End a call

*User:* kill the call to Mina, the agent's stuck
agent: call.list_active {}
→ [{ callSid: "CAxyz...", to: "+14155557890", durationSeconds: 142, billedUsd: 0.05 }]

agent: "I see one active call to +1 415 555 7890 — duration 2:22, $0.05 spent. End it?"
user: yes
agent: call.end { callSid: "CAxyz..." } → { ok: true }

Send an SMS follow-up

*User:* text Mina that the meeting moved to 2pm tomorrow
agent: contact.search { q: "Mina" }
→ [{ id: "c-1", name: "Mina Hart", number: "+14155557890" }]

agent: message.send { to: "+14155557890", body: "Hey Mina — meeting moved to 2pm tomorrow. — Sent via Voxrn" }
→ { messageSid: "SMabc...", segments: 1, billedUsd: 0.0079 }

agent: "Sent."

Troubleshooting

  • 401 Unauthorized from a Voxrn tool — your API key is wrong, expired, or the workspace it belongs to has been removed. Mint a fresh key at https://voxrn.com/enterprise/dashboard/agents → API keys.
  • Calls succeed but the agent says nothing — destination might have hit voicemail or returned caller_audio_failed. Check call.list_active for state. The Voxrn live monitor mirrors this view at https://voxrn.com/enterprise/dashboard/agents.
  • balance.check returns 0 — top up the workspace via Stripe or USDC at https://voxrn.com/account → Billing.

See also

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

94.39%
按下载量换算381

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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