AgentCall join-meeting skill

A skill that lets an agent join a Google Meet, Teams or Zoom call with voice, an avatar, screen sharing and live transcription

  • 194 ★ on the repository
  • Last commit
  • MIT
Skill

Broad access

Why this level

  • It connects to an external service and joins live meetings on your behalf
  • It can create an AgentCall account and spend credits, sending voice and screenshots to the cloud
All reasons and checks

pattern-ai-labs/agentcall

Install

Manual install

/plugin marketplace add pattern-ai-labs/agentcall
/plugin install join-meeting@agentcall

Also install dependencies: pip install aiohttp websockets.

Access and checks

How we review

Broad accessThe tool can change data, run commands or handle secrets.

  • It connects to an external service and joins live meetings on your behalf
  • It can create an AgentCall account and spend credits, sending voice and screenshots to the cloud

This is third-party code. Review the repository files before installing.

What it does

The skill connects a coding agent to a video meeting as a participant through the AgentCall service. The agent answers by voice via text-to-speech, receives transcripts of what is said, takes screenshots of a shared screen, shows an animated avatar instead of a camera and can present itself. It keeps its session context: it goes on searching code, editing files and running commands while talking in the meeting. The meeting becomes another input and output channel.

Who it is for. For developers and teams who want the agent present on a call by voice, not only in chat.

Good fit when

  • You want the agent to join a meeting and take part by voice
  • You want live transcription and screenshots of a shared screen during a call
  • You need the agent to present links, dashboards or slides

Not a fit when

  • You are not ready to create an AgentCall account and spend service credits
  • A bot's presence in the meeting is not agreed with participants
  • A text agent without voice and video is enough

Example request

Join this Google Meet call, follow the discussion and answer by voice when asked about the task status

Limitations

This is a skill on top of the AgentCall service, not a standalone app: it needs an account and credits, and new accounts have a free trial allowance. Voice, avatar and the meeting connection go through the service's infrastructure, so network access matters. It needs Python 3.10+ or Node.js 18+ and dependencies. Availability from Russia without a VPN is not guaranteed. Tell participants about the bot's presence. MIT licensed.

How to disable. Uninstall the join-meeting plugin or remove the skill folder from the agent's directory, and optionally delete the key from ~/.agentcall/config.json.

FAQ

Is this a standalone app?

No, it is a skill on top of a coding agent and the AgentCall service, running inside your agent.

Do I need a paid account?

You need an AgentCall account; new ones get free trial credits, then the service counts minutes.

How is AgentCall join-meeting different from meeting transcription bots?

The skill connects a coding agent to a video meeting as a participant through the AgentCall service. The agent answers by voice with speech synthesis, receives a transcript of lines and takes snapshots of the shared screen.

How do I install AgentCall and do I need a paid account?

In Claude Code: /plugin marketplace add pattern-ai-labs/agentcall, then /plugin install. It works in Cursor, Codex, Gemini CLI, Windsurf and Copilot. Meetings go through the AgentCall service, see its site for plan terms.

Discussion

No questions or reviews yet.

Skills

A set of everyday agent skills: web search, browser automation, Gmail, Google Calendar and Drive, speech transcription and YouTube transcripts

SkillExtended access2.7KRepository stars

Local AI commands for voice and text: clipboard dictation, text fixes, speech, an assistant and parallel agents in worktrees

CLIExtended access239Repository stars

Agent Standup shows Claude Code, Codex and Mistral Vibe sessions as a live team call: who works, who waits.

PluginMinimal access369Repository stars
Official

MeetCap records Mac meetings and transcribes them locally; its skill lets Claude Code and Codex record and search the archive

SkillBroad access305Repository stars

The /myoffice command opens a browser office showing what every agent on your machine is doing: thinking, waiting for you or done.

SkillExtended access401Repository stars

README in short

The README describes the join-meeting skill that lets a coding agent enter a video meeting as a bot and take part by voice. It lists features: text-to-speech speech, transcription, screenshots, an avatar and presenting during a call. Requirements are a coding agent, Python or Node.js with dependencies, and an AgentCall account that the agent can create itself via an email code. Install is shown for Claude Code, OpenClaw, Cursor, Codex, Gemini CLI and others through a plugin, an extension or git clone. MIT licensed.

README badge

Original SKILL.md

---
name: join-meeting
description: >
  AgentCall (agentcall.dev) — Join a video meeting (Google Meet, Teams, Zoom)
  as an AI bot with voice and visual presence. Supports audio-only mode with
  voice intelligence (barge-in, interruptions), text-to-speech mode, and
  webpage modes for custom UI. Use when asked to join a call, attend a meeting,
  or participate in a video conference.
argument-hint: <meet-url> [--mode audio|webpage-av] [--voice-strategy collaborative|direct] [--port PORT]
license: MIT
---

# join-meeting

Join a video meeting as an AI bot with voice and visual presence.

## Prerequisites

- Python 3.10+ (preferred) or Node.js 18+
- Python dependencies: pip install aiohttp websockets
- Node.js dependencies: cd scripts/node && npm install
- For webpage modes: a local HTTP server running on the specified port

Report a problem