voiceagent

SKILLAI 工具
AI 工具 未标注权限 许可证不明确
安装这个技能

Use this skill when the user wants to use an AI voice service for account login and authentication, creating dialogue flow agents or agentic prompt agents, creating outbound call tasks, associating TTS/ASR/dialogue agent IDs, uploading Excel contact lists with phone numbers and script variables, and configuring call result callbacks.

一句话选型结论

按资料字段整理

voiceagent是AI 工具类的 Agent Skill,安装方式未标注,权限级别未标注,许可证不明确,有可直接复制的安装命令。技能本质是给模型读的说明书,效果取决于模型与客户端;装完先用一个小任务验证,再放进正式流程。

分类AI 工具 安装方式未标注 权限级别未标注权限 许可证许可证不明确
累计安装 未标注 本周 未标注
GitHub 星标 未标注 仓库 未标注
下载量 未标注 技能市场下载数
权限级别 未标注权限 安装前请确认风险范围
许可证 许可证不明确 未说明

安装

未标注
在支持 Skills 的客户端里执行,装完问一句「你有哪些技能」验证
skillhub install user_a3035857/voiceagent
  1. 1
    确认客户端支持

    确认你的客户端支持 Skills(Claude、Codex、Cursor 等),并升级到较新的版本。

  2. 2
    拿到技能文件

    执行上面的命令安装;也可以把仓库里的 SKILL.md 手动放进客户端的 skills 目录。

  3. 3
    刷新技能列表

    重新打开客户端,或刷新技能列表,让新放进来的技能被加载。

  4. 4
    验证是否生效

    问一句「你有哪些技能」,确认列表里出现它,再拿一个小任务试跑。

技能档案

AI 工具
字段来自技能仓库与公开索引,中文说明由本站标注
作者 / 维护方user_a3035857仓库 未标注 · 版本 未标注
宿主客户端未标注技能可被这些客户端加载
安装方式未标注安装包 community
权限级别未标注权限安装前请确认风险范围
许可证未说明许可证不明确,商用前请自行确认条款
仓库可用性available技能仓库当前是否可访问
最近核验未标注上游最后核验时间 · 数据更新 1791624698
正文体量6.8 KBSKILL.md 原文大小,越大说明说明书写得越细

典型使用场景

3 类
按分类整理的常见用法,具体以 SKILL.md 为准
组合多个 AI 能力

把写作、检索、总结串成一条流程。

搭工作流

把固定流程做成可重复调用的步骤。

做提示词与评测

整理提示词并对比输出效果。

选择建议

装之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 有可直接复制的安装命令,不用自己翻文档找入口。

先想清楚的地方

  • 没有标注权限级别,实际能做什么要装好后确认。
  • 许可证不明确,商用或二次分发前先到仓库确认条款。
  • 没有做站内核验,内容以作者仓库为准。
  • 技能只是说明书,实际效果取决于模型与客户端,装完先用小任务试一遍。

更适合

  • 需要在AI 工具场景里给模型补一套做法的人。
  • 愿意先读一遍 SKILL.md、确认它会让模型做什么的人。

不太适合

  • 对许可证要求很严、必须明确可商用的人。
  • 只想装完就自动变强、不打算看说明书的人。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
技能权限级别安装方式累计安装宿主客户端安装命令
voiceagent当前 未标注权限 未标注 未标注 未标注 有
执行计划 只读 GitHub 仓库 151.5 万 Codex、Claude、Cursor 有
Playwright MCP 测试 浏览器控制 ClawHub 市场 136.5 万 OpenClaw 有
完成前的验证 未标注权限 GitHub 仓库 131.9 万 Codex、Claude、Cursor 有
子 Agent 驱动开发 只读 GitHub 仓库 131.2 万 Codex、Claude、Cursor 有
前端设计 执行命令 GitHub 仓库 125.6 万 Codex、Claude、Cursor 有

SKILL.md 全文

6.8 KB
模型实际读到的说明书原文,来自技能仓库

AI Voice Outbound

Use this skill when the user asks to create, configure, test, or manage AI voice outbound calling workflows.

Trigger examples include:

  • AI voice outbound calling
  • voice bot
  • outbound task
  • call campaign
  • dialogue flow agent
  • agentic prompt agent
  • TTS
  • ASR
  • upload call list
  • Excel contact list
  • callback URL
  • call transcript callback
  • Chinese-language requests about outbound call tasks, voice bots, dialogue agents, call-list uploads, or call-result callbacks.

Core Rules

  • Never invent account credentials, tenant IDs, tokens, callback URLs, agent IDs, TTS IDs, ASR IDs, task IDs, line group IDs, or contact list IDs.
  • Never start a live outbound task unless the user explicitly confirms the final task summary.
  • Treat real phone calls, production task starts, bulk contact uploads, and callback endpoint changes as high-impact actions.
  • Use a sandbox, staging, or test environment when the user is exploring or validating setup.
  • Validate Excel contact lists before upload.
  • If an API call fails, report the user-facing cause and the technical error code if available.
  • If API details are missing, continue with the workflow design and ask only for the missing API-specific fields needed for execution.

Capability Map

The skill supports five product capabilities:

  1. Account login and authentication.
  2. Voice agent creation, using either dialogue flow configuration or agentic prompt configuration.
  3. Outbound task creation, associating TTS, ASR, and voice agent IDs.
  4. Excel contact list validation and upload, including phone numbers and script variables.
  5. Callback configuration for outbound call results, including timing, duration, transcript, and structured call data.

End-To-End Workflow

When the user asks to create a complete outbound campaign:

  1. Authenticate or verify the active account.
  2. Collect missing business inputs from the user in plain language.
  3. Create or select a voice agent/process.
  4. Create an outbound task and associate the selected process and outbound line.
  5. Validate the Excel contact list.
  6. Upload the contact list.
  7. Configure the callback URL and callback fields.
  8. Summarize the final task configuration.
  9. Ask for explicit confirmation before starting live calls.

Natural Language Field Collection

Users usually describe the business goal instead of API fields. Do not ask them to provide raw field names unless they are technical users or the field has no business-friendly name.

When required values are missing, ask concise follow-up questions in business language.

For outbound task creation, collect:

  • Task name: ask "What should this outbound task be called?"
  • Dialogue process: ask the user to choose from available process names, then map to processCode.
  • Outbound line: ask the user to choose from available line names, then map to outboundNumber and outboundNumberId.
  • Task type: ask whether calls should be manual or scheduled, then map to taskType.
  • Start time: ask only when the task is scheduled/automatic.
  • Concurrency: ask "How many calls can run at the same time?", then map to lineAmount.
  • Duplicate-phone rule: ask whether duplicate phone numbers are allowed, then map to permitRepeatNum.

If a required value can be discovered from an API, fetch candidates and present the names instead of asking the user for IDs or codes.

Before calling a creation API, summarize the mapped fields and proceed only if the missing required inputs have been resolved.

1. Authentication

Before using protected APIs, authenticate or verify the active session.

Required inputs:

  • Environment: production, staging, sandbox, or private deployment.
  • Account, tenant, or workspace identifier.
  • Login method.
  • Credential source or existing access token.

Return:

  • Authenticated account identity.
  • Tenant or workspace ID.
  • Token/session status.
  • Expiration time, if available.

Read references/auth.md when implementing or using API-specific authentication.

2. Voice Agent Creation

Support two creation modes:

  • flow: create or update a structured dialogue flow.
  • agentic_prompt: create or update a prompt-driven voice agent.

Required inputs:

  • Agent name.
  • Scenario and business goal.
  • Creation mode.
  • Opening message.
  • Required variables to collect.
  • Fallback behavior.
  • Transfer, retry, or hangup rules.

Return:

  • agent_id.
  • Agent name.
  • Creation mode.
  • Status.

Read references/agent.md when mapping user requirements to agent configuration.

3. Outbound Task Creation

Create an outbound task after the agent, TTS, and ASR resources are known.

Required inputs:

  • Task name.
  • agent_id.
  • TTS voice or resource ID.
  • ASR model or resource ID.
  • Call time window.
  • Concurrency or rate limit.
  • Retry policy.
  • Caller number, line group, or routing resource.

Return:

  • task_id.
  • Task status.
  • Associated resource IDs.

Read references/outbound-task.md for the task schema and lifecycle.

4. Excel Contact List Upload

Validate the contact list before upload.

Required column:

  • phone_number

Common optional columns:

  • customer_name
  • city
  • product_name
  • due_date
  • Custom variables used by the voice agent.

Validation checklist:

  • Required columns exist.
  • Phone numbers are valid for the expected region.
  • Empty rows are ignored or flagged.
  • Duplicate numbers are reported.
  • Variable column names match the agent's expected variables.
  • Row count and upload limits are respected.

Read references/contact-upload.md before validating or uploading contact files.

5. Callback Configuration

Ask the customer for a callback URL before enabling result push.

Recommended callback fields:

  • task_id
  • call_id
  • phone_number
  • start_time
  • end_time
  • duration_seconds
  • call_status
  • hangup_reason
  • transcript
  • extracted_variables
  • recording_url

Read references/callback.md when configuring or testing callbacks.

Missing API Details

The API contract is intentionally kept in reference files. If the implementation details are not available yet, produce a configuration plan and clearly mark unresolved fields as TBD.

Do not block the entire workflow only because API endpoint paths or request schemas are missing.

API References

For the full SaaS API map, read references/saas-api-overview.md.

For raw extracted text from the source Word document, read references/saas-api-extracted.txt only when the overview and focused reference files do not contain enough detail.

常见问题

5 问
答案由本站依据条目字段整理
voiceagent怎么安装?

在支持 Skills 的客户端里执行:skillhub install user_a3035857/voiceagent。安装方式为未标注。装完刷新技能列表或重开客户端。

voiceagent支持哪些客户端?

资料没有标注宿主客户端,但只要客户端支持 Skills(Claude、Codex、Cursor 等),一般都能加载。

voiceagent权限高吗,安全吗?

权限级别为未标注权限。安装前先看它会读取或写入哪些内容。

voiceagent收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于技能本身和所用的模型或客户端,正式使用前按作者说明确认。

voiceagent用什么许可证?

许可证标注为 未说明(许可证不明确)。商用或二次分发前请到仓库确认条款原文。

热门专题

按场景挑技能
全部专题 →

技能内容以作者仓库为准;分类、权限级别与许可证状态的中文说明为中文整理,最近核验 未标注。