Google Sheets MCP Setup

Live research data is stored in Google Sheets (source of truth). Cursor agents read and write via workspace-cpalss — full setup: Google Workspace MCP.

Sheet registry

Canonical links live in sheets-registry.json.

Dataset	Spreadsheet
LNY market landscape	Open in Google Sheets
Autumn market landscape	Open in Google Sheets

LNY spreadsheet ID: 1IB6beiKHC9R05r2o0imkh8MX_2u-qmhlVxTV3pxibs0
Autumn spreadsheet ID: 1Va0oCv09nyW98-JnWL9DuYYTsXG35cxjf_WqBZsySrA

Shared Drive folder layout

Recommended structure under your Festival Network Shared Drive:

Operations/Festival Network/
  Research/       ← market landscape, market analyses (use this, not Marketing)
  Calendars/      ← future: roll-up paths, scheduling
  Graphs/         ← future: ecosystem edges
  Operations/     ← future: steering trackers

Move the market landscape spreadsheet from Marketing to Research — that sheet is coalition market research, not outbound marketing collateral.

Cursor / agent access

Use workspace-cpalss. Setup and auth: Google Workspace MCP. Discover exact tool names with Cursor’s MCP tool list (GetMcpTools) — official Google Workspace MCP names differ from the retired @dguido/google-workspace-mcp tools.

Script OAuth for Node pipelines (not Cursor MCP) still uses .google-workspace-oauth.json via npm run mcp:oauth -w festival.

Drive/Docs notes: google-drive-mcp.html.

Operating rules

• Sheets = source of truth — humans edit in Google Sheets; agents update via workspace-cpalss
• Repo holds sheets-registry.json, column schema, and guides — not live rows (no CSV)
• Match columns in LNY schema or Autumn schema
• Protect row 1 (headers) in the Sheet