What is Panacea?

Panacea is an AI-powered knowledge management and customer support platform that lets you build a living knowledge base, train an AI on your business, and deliver instant, accurate answers to your customers — around the clock.

What Panacea does

At its core, Panacea connects three pieces:

1. Your knowledge — Ingest documentation, URLs, policies, FAQs, and any written content into a searchable, AI-indexed wiki.
2. An AI agent — Automatically answers customer questions by retrieving the most relevant knowledge, with citations back to the source entries.
3. A human layer — When the AI can't confidently answer, it escalates to your team in real time via the Inbox.

Who is it for?

Panacea is designed for:

• SaaS companies that want to reduce support ticket volume without sacrificing quality.
• Developers and product teams that need a private, hosted knowledge base for internal or external documentation.
• Customer success teams that want visibility into what questions are being asked, how well the AI is performing, and where knowledge gaps exist.

Core concepts

Concept	What it means
Wiki entry	A single knowledge base article — the unit of truth your AI uses to answer questions.
Ingest	The process of adding new content (URL, raw text, file) to your knowledge base.
Query	A question asked to the AI agent — it retrieves relevant wiki entries and synthesises an answer.
Review queue	AI-proposed updates to wiki entries that need a human to approve or reject before going live.
Escalation	A session the AI couldn't resolve, routed to a human agent via the Inbox.
Source	An external integration (e.g. Airdraft CMS, GitHub) that syncs content into the wiki automatically.
Session	A conversation thread between a customer and the AI agent (or a human agent during escalation).

The workflow at a glance

Content → Ingest → Wiki → AI Agent → Customer query
                              ↓              ↓
                        Review queue    Escalation (Inbox)
                              ↓              ↓
                         Approved       Human agent
                              ↓
                       Updated wiki

Next steps

• Set up your account and first project
• Add content to your knowledge base
• Test your AI agent in the Query playground
• Embed the chat widget on your site