VCONIC team conversations as vCons

Team Google Meet meetings and business email, privacy filtered, as IETF vCon records your agent can search. Internal, read-only, token required.

Endpoint https://team.vconic.com/mcp
Token Bearer <team token>

~930 meetings~440 emailsas of 2026-09-23, growing every ten minutes

The token is not on this page. Ask the service owner for it, keep it out of repos, chats and screenshots, and replace <team token> below with it.

What a vCon is

A vCon is one conversation as a single JSON record, the IETF's container format for recorded conversations. Every record here is one meeting or one email. It has four parts: parties (who took part), dialog (the conversation itself: the email text), analysis (things derived from the dialog) and attachments (documents about the conversation). A meeting carries its Gemini notes and transcript as a meeting_document attachment. Every record also carries a team_filter attachment describing the privacy review and a lawful_basis attachment. Trimmed example:

one meeting vCon, abridged
{
  "subject": "Filtered team meeting document",
  "parties": [ { "name": "Meeting participants (identities withheld)" } ],
  "attachments": [
    { "purpose": "meeting_document", "mediatype": "text/plain",
      "body": "VCONIC Stand-up - Notes by Gemini\n\nNotes ... Transcript ..." },
    { "purpose": "team_filter", "encoding": "json",
      "body": "{\"source_kind\": \"drive\", \"derived\": true, ...}" },
    { "purpose": "lawful_basis", "encoding": "json",
      "body": "{\"lawful_basis\": \"legitimate_interests\", ...}" }
  ]
}

So "what was said in a meeting" means the meeting_document attachment, and "what an email said" means the email's dialog. Names, email addresses, phone numbers and personal attacks are removed before a record is published; business feedback and customer complaints are kept. Email parties appear as Participant 1, Participant 2 and so on.

Claude Code, one line

terminal
claude mcp add --transport http vcon-team https://team.vconic.com/mcp --header "Authorization: Bearer <team token>"

Cursor, Windsurf, Cline, any client that sends headers

mcp config JSON
{
  "mcpServers": {
    "vcon-team": {
      "type": "http",
      "url": "https://team.vconic.com/mcp",
      "headers": { "Authorization": "Bearer <team token>" }
    }
  }
}

claude.ai and ChatGPT connectors

Not supported. Those connectors cannot send a header, and this data has no open endpoint. Use Claude Code, another header-capable client, or the REST API.

Plain REST

shell
curl -s -H "Authorization: Bearer <team token>" --get \
  --data-urlencode "q=pricing" --data "limit=5" \
  https://team.vconic.com/api/v1/vcons/search/content

Try asking your agent

About the data

Meetings are Google Meet events organised from a strolid.com or vconic.com address whose attached Gemini document has a Transcript tab. Email is business correspondence between an employee and anyone else; personal, HR, legal, medical, newsletter and automated mail is excluded. Each record is a filtered copy; originals stay in Google. Ingestion is still backfilling history, so older meetings and mail are incomplete. Server: vcon-mcp. Ingestion: VCONIC/vconic-team-service (private).