Google Drive lands — the connector pack is complete
The Drive connector — Docs, Sheets, Slides, and text files exported into scoped, retrievable knowledge — completes the connector set the two reference deployments run on.
The last connector in the pack is in (v1.35): Google Drive. With it, the set we set out to build
for the two reference deployments is complete.
What it brings in
Drive is where the narrative work of an agency lives — proposals, briefs, case-study and proof docs.
The connector lists files (incremental by modifiedTime, resumable from a cursor, scoped to a query
or folder) and pulls real text: Google Docs and Slides export as plain text, Sheets as
CSV, and plain-text files download directly. Binary files yield metadata only — no junk in the index.
Like every connector, it rides the durable, incremental pipeline: a sync fetches only what changed, a failure resumes from the last checkpoint, and ingested docs inherit the client scope so one workspace serves many clients without leakage.
The pack, complete
HubSpot · CRM contacts / deals / companies (RevOps)
Gmail · messages (RevOps)
Slack · channel history (RevOps)
Google Drive · docs / proposals / proof (RevOps)
Google Ads · campaign performance by day (Daylyte)
GA4 · sessions / conversions / CRO (Daylyte)
That's exactly the surface our two reference deployments need: HubSpot + Gmail + Slack + Drive for Metacto RevOps, Google Ads + GA4 for Daylyte's PPC/CRO reporting. The platform can now touch the real systems the work happens in — which is the whole point of a runtime over a demo.
Where this leaves the road to 1.0
The engine, the control plane (REST + MCP-over-HTTP), and the connector pack are all in. What's left is the satisfying part: standing up the reference deployments themselves — wiring these connectors, running the missions, routing the approvals — and hardening from what real work surfaces. That's next.