Knowledge Base
XPT Tracker has no separate wiki. It has one Markdown engine that hosts every kind of content as a hierarchical page tree. What differs between pages is the kind of content; the structure and editor are the same.
| Content | Where you see it | Purpose |
|---|---|---|
| Docs | Sidebar → Docs | Structured, shared documentation |
| Notes | Sidebar → Notes | Lighter, free-form pages |
| Learnings | Sidebar → Context | Reusable patterns agents capture from past work — each with its track record |
| Artifacts | Attached to tasks | The documents workflows produce: plans, results, learnings per task |
Why one engine, many kinds: knowledge usually scatters across README, wiki, and chat — each at a different durability. XPT Tracker keeps it in one system with one search, one format, and one ownership model (the workspace). That also means agents read and write the same knowledge base you do — a task's plan, a decision, a reusable pattern all live here.
The processes agents follow, and the templates their artifacts are made from, live in AI Workflows in the sidebar.
This section covers:
- Docs & Notes — the editor, the page tree, and version history.
- Context — the knowledge your agents accumulate, and whether it helps.
- Publishing & Visibility — who can read a page.