Logo
ClawHub Skills Lib
HomeCategoriesUse CasesTrendingBlog
HomeCategoriesUse CasesTrendingBlog
ClawHub Skills Lib
ClawHub Skills Lib

Browse 20,000+ community-built AI agent skills for OpenClaw. Updated daily from clawhub.ai.

Explore

  • Home
  • Trending
  • Use Cases
  • Blog

Categories

  • Development
  • AI & Agents
  • Productivity
  • Communication
  • Data & Research
  • Business
  • Platforms
  • Lifestyle
  • Education
  • Design

Use Cases

  • Security Auditing
  • Workflow Automation
  • Finance & Fintech
  • MCP Integration
  • Crypto Trading
  • Web3 & DeFi
  • Data Analysis
  • Social Media
  • 中文平台技能
  • All Use Cases →
© 2026 ClawHub Skills Lib. All rights reserved.Built with Next.js · Supabase · Prisma
Home/Blog/Google Meet: Create and Manage Video Conferences Directly from Your AI Agent
skill-spotlightapi-standardsgoogle-meetclawhubopenclawgoogle-workspace

Google Meet: Create and Manage Video Conferences Directly from Your AI Agent

March 12, 2026·6 min read

15,409 downloads, 13 stars, 6 installs. The Google Meet skill by @byungkyu brings the Google Meet REST API into your Claude Code workflow with zero OAuth complexity. Create meeting spaces on demand, pull conference records, manage participant access — all through a single MATON_API_KEY that handles Google authentication for you.

The Problem It Solves

The Google Meet API requires OAuth 2.0 with specific scopes, a Google Cloud project, client credentials, token refresh handling, and — if you want calendar integration — additional Calendar API setup. For most developers, this is a two-hour project before they can make their first API call.

byungkyu's Maton-based skills solve this with a gateway pattern: connect your Google account once at ctrl.maton.ai, and from that point forward all API calls go through gateway.maton.ai/google-meet/... with your MATON_API_KEY. The gateway handles OAuth token injection, refresh, and Google's scope management. You never touch a Google Cloud console.

How the Maton Pattern Works

Every byungkyu skill on ClawHub follows this architecture:

Your Agent
    │
    ▼
gateway.maton.ai/google-meet/{endpoint}
    │  (Authorization: Bearer MATON_API_KEY)
    ▼
meet.googleapis.com/{endpoint}
    │  (OAuth token auto-injected by gateway)
    ▼
Google Meet API Response

The gateway proxies your request to the native Google Meet API endpoint (meet.googleapis.com) and injects your connected Google account's OAuth token. From the Google API's perspective, it's an authenticated first-party request. From your agent's perspective, it's a simple bearer token call.

Core Capabilities

Create a Meeting Space

python <<'EOF'
import urllib.request, os, json
 
data = json.dumps({}).encode()
req = urllib.request.Request(
    'https://gateway.maton.ai/google-meet/v2/spaces',
    data=data, method='POST'
)
req.add_header('Authorization', f'Bearer {os.environ["MATON_API_KEY"]}')
req.add_header('Content-Type', 'application/json')
 
response = json.load(urllib.request.urlopen(req))
print(f"Meeting URL: {response['meetingUri']}")
print(f"Space ID: {response['name']}")
EOF

This creates a new Google Meet space and returns the meeting link immediately. The link is persistent — it can be reused for recurring meetings.

List Conference Records

python <<'EOF'
import urllib.request, os, json
 
req = urllib.request.Request(
    'https://gateway.maton.ai/google-meet/v2/conferenceRecords'
)
req.add_header('Authorization', f'Bearer {os.environ["MATON_API_KEY"]}')
 
records = json.load(urllib.request.urlopen(req))
for record in records.get('conferenceRecords', []):
    print(f"Conference: {record['name']}")
    print(f"  Start: {record.get('startTime')}")
    print(f"  End: {record.get('endTime')}")
EOF

Returns historical meeting records — useful for attendance tracking, billing, or compliance documentation.

Get Meeting Participants

python <<'EOF'
import urllib.request, os, json
 
conference_id = "conferenceRecords/abc123"
req = urllib.request.Request(
    f'https://gateway.maton.ai/google-meet/v2/{conference_id}/participants'
)
req.add_header('Authorization', f'Bearer {os.environ["MATON_API_KEY"]}')
 
participants = json.load(urllib.request.urlopen(req))
for p in participants.get('participants', []):
    print(f"Participant: {p.get('signedinUser', {}).get('displayName', 'Anonymous')}")
    print(f"  Joined: {p.get('earliestDeviceJoinTime')}")
EOF

Manage Multiple Google Accounts

If you need to create meetings across different Google Workspace accounts (e.g., different clients):

python <<'EOF'
import urllib.request, os, json
 
req = urllib.request.Request(
    'https://ctrl.maton.ai/connections?app=google-meet&status=ACTIVE'
)
req.add_header('Authorization', f'Bearer {os.environ["MATON_API_KEY"]}')
connections = json.load(urllib.request.urlopen(req))
# Use Maton-Connection header to specify which account
EOF

Setup

clawhub install google-meet

Connect your Google account:

  1. Sign up at maton.ai and get your MATON_API_KEY
  2. Go to ctrl.maton.ai
  3. Create a new Google Meet connection — it will open Google OAuth in your browser
  4. Authorize and return
export MATON_API_KEY="your_maton_key"

Access Transcripts and Smart Notes

One of Google Meet's most powerful features for AI workflows: native meeting transcripts and smart notes are accessible via the API.

python <<'EOF'
import urllib.request, os, json
 
conference_id = "conferenceRecords/abc123"
 
# Get transcripts
req = urllib.request.Request(
    f'https://gateway.maton.ai/google-meet/v2/{conference_id}/transcripts'
)
req.add_header('Authorization', f'Bearer {os.environ["MATON_API_KEY"]}')
transcripts = json.load(urllib.request.urlopen(req))
 
# Get individual transcript entries (full text with timestamps)
for transcript in transcripts.get('transcripts', []):
    entries_req = urllib.request.Request(
        f'https://gateway.maton.ai/google-meet/v2/{transcript["name"]}/entries'
    )
    entries_req.add_header('Authorization', f'Bearer {os.environ["MATON_API_KEY"]}')
    entries = json.load(urllib.request.urlopen(entries_req))
    for entry in entries.get('transcriptEntries', []):
        print(f"[{entry.get('startTime')}] {entry.get('participant')}: {entry.get('text')}")
EOF

Transcripts include full text with timestamps and speaker attribution. Combined with Claude, this enables automatic meeting summarization, action item extraction, and decision documentation — all triggered by the recording ready webhook without any human intervention.

Note: transcripts and recordings are generated asynchronously after a meeting ends, typically within a few minutes.

Use Cases

Automated meeting scheduling pipelines. When a booking is confirmed (via Calendly, your CRM, or a form), have Claude automatically create a Google Meet space and return the link — no human needs to click "New Meeting."

Conference record auditing. Pull all meeting records for a date range, extract participant lists and durations, generate attendance reports. Useful for compliance (education, healthcare, legal) where meeting documentation is required.

Support ticket escalation. When a support ticket hits a certain threshold, Claude creates a Meet space and sends the link to the customer — on-demand video support without manual coordination.

Team standup automation. Create recurring meeting spaces with consistent URLs, archive participant records after each session, log attendance to a spreadsheet or database.

Virtual event coordination. Programmatically create breakout rooms, track which sessions participants joined, generate post-event attendance reports.

Practical Tips

Meeting spaces are reusable. A created space has a persistent meetingUri. For recurring meetings (weekly syncs, office hours), create the space once and reuse the same link. No need to generate a new link each time.

Conference records lag slightly. Google Meet conference records aren't available instantly after a meeting ends. If you're building automation that processes meeting records, add a delay (15–30 minutes) before querying for new records.

Combine with the Calendly skill. byungkyu also publishes a Calendly API skill. A common pattern: when someone books a Calendly event, a webhook triggers Claude to create a Google Meet space and update the Calendly event with the meeting link. Both skills use the same MATON_API_KEY.

Considerations

  • Read-only participant data. The Google Meet API supports listing participants but not forcibly removing them or muting them — those actions require the Meeting Admin role and in-meeting controls.
  • Only 6 installs reflects niche use. The skill is highly useful for specific workflows, but most Google Meet creation happens through Google Calendar integration rather than direct API calls. This skill is for programmatic/agentic use cases, not replacing the UI.
  • Google Workspace accounts work best. The Meet API has fuller functionality with Google Workspace (Business/Enterprise) accounts compared to personal Gmail accounts.
  • MATON_API_KEY is a shared credential. The key grants access to all your connected accounts on Maton. Treat it with the same care as a service account key.

The Bigger Picture

The Google Meet skill is part of a broader pattern in byungkyu's catalog: wrapping Google Workspace, Calendly, ClickUp, and other enterprise APIs through the Maton gateway so they're accessible to AI agents without OAuth boilerplate. Individually, each skill handles one specific service. Together, they form a suite that lets Claude participate in enterprise workflows that previously required humans at every API integration point.

Creating a meeting, inviting participants, tracking who attended, logging it to a project management tool — this chain of tasks maps directly to byungkyu's Google Meet + Calendly + ClickUp skill combination.

View the skill on ClawHub: google-meet

← Back to Blog