Google Docs / Sheets / Gmail / CalendarFirst met in aicp-c3-basecamp-03 · Productivity & source apps
The everyday apps automation reads from and writes to — and the first connectors anyone wires up.
You need a destination that a non-technical person can already open. A Sheet is a database everyone can read.
Universally available as both MCP connectors and n8n nodes
Sheets works as a stand-in CRM — BC6's post-call data landed there with drag-and-drop column mapping
Docs as a workflow terminal: create and append, verified live in BC3
Receives scraped and generated content (Docs create/update, Sheets append-row)
Supplies calendar and mail context to assistants via connectors
Acts as the human-readable end of an automated pipeline
Doesn't scale as a database — a Sheet is a stand-in, not a system of record
Doesn't send bulk mail safely through the Gmail node
The endpoints of most workflows in the course: read from Calendar and Gmail, write to Docs and Sheets.
Productivity & source apps
The pipeline's outputs are local files and static HTML.
A Sheet mirroring extraction progress is the cheapest possible dashboard if Paul ever wants to check status without opening the site.
current
n8n: native nodes for all four
claude-desktop-mcp: Gmail and Calendar are the flagship connectors
calcom: bookings land in Calendar
| Session | Coverage | Moment | What happened |
|---|---|---|---|
| Basecamp 3: Introduction to n8n — I | demonstrated | 2:46:31 | Terminal node of the demo workflow — doc created and scraped data appended, verified live |
| Basecamp 4: Introduction to n8n — II | demonstrated | 1:45:01 | Create + update document nodes, drive folder selection, expression-based titles |
| Basecamp 4: Introduction to n8n — II | demonstrated | 1:53:09 | Send-message delivery node; hardcoded-vs-expression subject discussion |
| Basecamp 6: Voice Agents | demonstrated | 2:45:20 | Stand-in CRM; columns mirror post-call properties; document by ID/URL |
| Basecamp 6: Voice Agents | demonstrated | 2:21:00 | Connected through cal.com; live booking landed in it during the test call |
| Session 6: Building an AI-Powered Lead Gen Machine | demonstrated | 1:17:21 | The factory spine: row-added trigger, intermediate logging ('before you lose the data'), context columns appended per station |
| Session 17: Facebook Ads Competitor Research & Replication | demonstrated | 1:35:29 | Two roles: hosting the product image (where the share-link HTML-viewer trap surfaced) and receiving generated ads for the human-review gate |
| Session 14: Claude Code — Advanced Workflows | mentioned | 0:35:48 | 'The most used CLI' — full Google Workspace control; promised for a later session |
| AI Sprint: AI Evals & Reliability — Day 2 (Testing Systems That Never Answer Twice) | mentioned | Where the audience agent writes its drafts |