Your AI now schedules, replies and sets up your company from its chat

ōrbita's connector for Claude, ChatGPT and other assistants grows: with your permission and your confirmation in the chat, your AI schedules or publishes a post and replies to comments, mentions and Google reviews; reads and edits the Brand Kit, reads the interactions inbox and fills in the settings of a new company with your documents. All without leaving its chat and with the same rule always: nothing is sent unless you ask for it and confirm it.

Updated Author: Mónica Lorenzo

Areas: Your connected AIContentInteractionsBrand Kit and data

What changes

Until today, your AI assistant left drafts in ōrbita and you took the last step in the app. From today it can finish the job without you leaving its chat, when you ask it to:

  • Schedule or publish a post. It shows you the exact text, the networks and the date; when you say yes, it schedules it (or publishes it right away). One post per confirmation, never in bulk. TikTok is left out: TikTok requires that you are the one who chooses the privacy and sees the preview, so your AI explains it to you and gives you the link to finish it in ōrbita.
  • Reply to comments, mentions and Google reviews on behalf of your company, with the same rule: it shows you the reply and sends it only when you confirm. Private messages are left out.
  • Read the interactions inbox of your networks to prepare those replies.
  • Read and edit the Brand Kit: colors, typefaces, hashtags, notes and material links. You still upload the logos yourself.
  • Set up a new company. You hand it the catalog, a dossier or the website, and your AI fills in the website, answers the questions about the business, adds products, frequently asked questions and dates, and asks for the communication plan. You confirm the plan, in ōrbita or in the chat; afterwards it can ask for the first month of content.

Scheduling, publishing and replying are new, optional permissions: they come ticked like the others and you untick them if you do not want them. Without them, everything stays as before. And with them, nothing goes out unless you ask for it and confirm it in the chat.

What you need to do

If you already had your AI connected, connect it again once to see the new permissions and choose which ones you grant. In ChatGPT, also remove the connector and create it again: it keeps the list of features from the day you added it. The steps are in Connect your AI.

What ōrbita records

For each confirmed action, ōrbita keeps for 90 days the instruction it executed (text, network and date) and that your assistant declared your confirmation. Never your conversation. We have updated the terms (version 1.9) and the privacy policy (version 1.18) to reflect it.

All releases

Published . Updated .