Show HN: 우리는 Claude Code가 실제 팀원처럼 회의에 참여할 수 있는 방법을 마련했습니다

hackernews | | 🖥️ 하드웨어
#ai os #anthropic #claude #conway #review #에이전트 #클로드 코드 #하드웨어/반도체
원문 출처: hackernews · Genesis Park에서 요약 및 분석

요약

개발자가 터미널에서 실행하는 코딩 에이전트가 구글 미트나 줌 등의 회의에 실시간으로 참여할 수 있는 ‘AgentCall’이 출시되었습니다. 이 도구는 기존 세션과 파일 접근 권한을 유지하며, 낮은 지연 시간의 음성 대화와 로컬 웹페이지 화면 공유가 가능합니다. 또한 에이전트가 직접 마이크를 조절하거나 발언을 요청하는 등 실제 팀원처럼 상호작용할 수 있는 기능을 제공합니다.

본문

Hi HN — we built AgentCall because the coding agent you&#x27;re already running in your terminal shouldn&#x27;t be trapped there.<p>The pitch: your existing Claude Code, Codex, OpenClaw, or Cursor session joins a Google Meet, Teams, or Zoom call as itself. Same session, same context, same file access. It speaks, listens, screen-shares a localhost webpage, and can code live while you all talk about what it&#x27;s building.</p><p>What&#x27;s actually on the call:</p><p>• Voice in, voice out. Two modes — collaborative (sub-second via a voice intelligence layer tuned for latency) or direct (~2s, your coding agent itself doing the talking, with full reasoning).</p><p>• Screen share that is <i>not</i> a desktop grab. It&#x27;s a URL or local port rendered into the meeting through a per-call tunnel (temp subdomain + secret token, dies when the call ends). The agent chooses what to expose, nothing else leaks.</p><p>• Shareable webpage links that participants open in their own browser — live dashboards, diffs, forms — served from the agent&#x27;s localhost, tunneled only for the call.</p><p>• Meeting chat: the agent reads incoming messages (great for dropping in URLs, error logs, code snippets that sound terrible over TTS) and sends them back — your agent can paste a PR link while still talking.</p><p>• Participant awareness: the agent knows who&#x27;s in the room, who joined, who left, and who&#x27;s actively speaking. It can raise its hand to speak politely, toggle its own mic, and change voice&#x2F;interruption&#x2F;barge-in settings on the fly.</p><p>• Leave and rejoin on command. Ask mid-call (&quot;leave for two minutes and come back&quot;) and the agent does it.</p><p>• TS&#x2F;STT run on our own servers (several voices, English STT today, multi-language TTS). No data to third parties in direct mode.</p><p>Everything above is controlled by your coding agent through the skill. There are no hardcoded meeting behaviors — the agent decides when to speak, when to chat, when to raise a hand, what to share.</p><p>What we don&#x27;t do: we don&#x27;t run the coding agent. The agent&#x27;s model calls, file I&#x2F;O, tokens, and tools all happen on your machine or your cloud. We don&#x27;t store recordings or screen captures by default — everything streams to the agent in real time and the agent decides what to keep. Transcripts are in-memory for crash resilience and wiped on disconnect unless you opt into retention (up to 7 days).</p><p>We bundle TTS + voice + tunneling, start at $0.35&#x2F;min (drops with volume), and store nothing by default. We join existing meetings — we&#x27;re not a new video platform.</p><p>Get AgentCall Running — 2 Minutes
Works with Claude Code (best), Codex, and Cursor.</p><p>1. Install the skill
install join-meeting skill from <a href="https:&#x2F;&#x2F;github.com&#x2F;pattern-ai-labs&#x2F;agentcall" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;pattern-ai-labs&#x2F;agentcall</a></p><p>2. Get your API key
Sign up at <a href="https:&#x2F;&#x2F;agentcall.dev" rel="nofollow">https:&#x2F;&#x2F;agentcall.dev</a> , copy your key from the dashboard, and paste it when the skill asks.</p><p>3. Invite the agent to a meeting
Paste any Google Meet &#x2F; Zoom &#x2F; Teams URL into your agent:</p><p>Give it 30–60 seconds for the bot to spin up. Once it joins, start talking.</p><p>Demo: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;@pattern-ai-labs" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;@pattern-ai-labs</a></p><p>Would love feedback on the voice latency in direct mode, the privacy model, and where the &quot;joining existing meetings&quot; framing breaks for your workflow.</p>

Genesis Park 편집팀이 AI를 활용하여 작성한 분석입니다. 원문은 출처 링크를 통해 확인할 수 있습니다.

공유

관련 저널 읽기

전체 보기 →