
Your people already have a favourite AI assistant. From today, they can use your company’s knowledge in Augela from inside it: Claude on the web, desktop and mobile, ChatGPT where the plan allows custom connectors, Microsoft Copilot Studio, Claude Code, and any other assistant that supports MCP, the open standard these assistants use for connectors.
They ask your company’s agents and search its confirmed knowledge from where they already work, and everything stays under your company’s governance in Augela.
What someone can do from their assistant
- See your agents: the ones they may ask, with what each one is for.
- Search your company’s knowledge: confirmed facts first, with who confirmed them and when, then documents with their sources.
- Ask an agent: its full governed answer with sources, as in Augela itself. Follow-up questions continue the conversation.
- Report a wrong answer: it goes to your Review Queue like a thumbs-down in Augela. The reporter later sees what came of it.
Fixed once, right everywhere
Here is a real example from testing. Someone asked their assistant about expedite times, and Augela answered from confirmed memory: 5 days. They knew it was 14 days now, so the assistant reported the answer. A reviewer corrected it in Augela, once. The next answer, in any assistant, was 14 days, “confirmed today”.
That is company memory and wisdom at work: what your people confirm is used first, wherever the question comes from.
Governed like everything else
- Off until you say so. A Company admin turns it on and confirms that answers go to the assistant that asked, under your company’s agreement with that vendor.
- Your access rules. Admins choose which agents can be asked, per Company, Group or agent. Each person sees only what their own access allows.
- Checked first. Every question meets your watch words before anything is searched. In Proactive mode, a flagged answer is held for review and never sent.
- Limits and a record. Limits per person and per company, within your AI budget, and every connection, call, refusal and report in the audit trail.
- Visible. Answers to assistants show as their own channel, “AI assistants”, in the Governance Health Report.
How to connect
In Claude, add a custom connector with https://interface.augela.com/api/mcp, sign in with Augela and allow access. No key needed. ChatGPT uses the same address where your plan allows custom connectors; Copilot Studio, Claude Desktop, Claude Code and scripts use a key created in Augela. The full steps are in the connection guide.
Both directions
Augela agents already looked things up live in your own systems through MCP. Now the big assistants can ask Augela too. Your company’s memory and rules stay in one governed place, and your people use whichever assistant they prefer. It is included in every plan.
Read more on Augela in your AI assistants.
Want to see it on your own content? Start free, no card required.

