OpenHuman Guide

OpenHuman Guide

Desktop setup & workflows · Updated

Start setup

Guide

First prompts after sync

  • · Only run these after Gmail (or whichever anchor connector) clears at least one auto-fetch lap (~20 minutes).
  • · Replace placeholder names (“Sarah”) with real teammates you actually email or DM.
  • · If answers feel skinny, revisit Integrations & sync before swapping models.

Goal

Prove the assistant remembers your inbox or channels: you should see references to real threads, not generic ChatGPT filler. If everything still sounds hollow, treat it as “not enough data synced yet.”

Start here (copy/paste)

Paste one at a time. Watch for specifics (senders, ticket numbers, Slack channels). No specifics yet? Wait for sync, then retry.

  1. "What's waiting on me?"

    Expect: If sync already ran: actionable bullets from mail/chat. If not: vague or empty—go back to Integrations and wait.

  2. "What do I need to know from the last 12 hours?"

    Expect: Good sanity check once mail or chats have ingested; weak output usually means connectors are still cold.

  3. "Summarize what I missed today."

    Expect: Mixes threads across sources once each connector finished its first ingest pass.

Briefings (Getting Started mirror)

Once memory is warm, these match upstream Getting Started flavor:

  • "What do I need to know from the last 12 hours?"
  • "What's waiting on me?"

Cross-source questions

  • "Summarize what I missed today."
  • "What are the key decisions from this week?"
  • "Extract action items from my recent conversations."
  • "What did Sarah say about the project across email and chat?"

These assume multiple connectors finished at least one ingest pass. Still thin? Integrations & sync before blaming the model.

Optional: voice, meetings, advanced surfaces

Once basics work, upstream docs describe meeting agents, native voice, and background loops—each adds permissions and cost. Opt in deliberately.