mrge-io Codebase Context
Query cubic's AI-generated wiki to answer questions about codebase architecture and design.
When To Activate
- User asks how a system or feature works
- User is onboarding to a new codebase
- User needs to understand how components connect
- User asks about architecture, data flow, or system design
- User is making changes that span multiple modules and needs context
Inputs
- Question (required): What the user wants to know about the codebase.
Instructions
1. Detect the repository
git remote get-url origin
Extract the owner and repo name from the remote URL.
2. Find relevant pages
Call list_wiki_pages with the owner and repo. Match page titles and descriptions to the user's question.
3. Fetch content
Call get_wiki_page for the most relevant pages (1–3 pages max).
4. Synthesize an answer
Combine wiki content into a focused answer for the user's specific question. Connect wiki knowledge to the specific files or modules the user is working with.
Output Format
Summarize the relevant wiki content — don't dump entire pages. Reference specific wiki pages so the user can explore further.
Based on the cubic wiki for this repo:
**Authentication Flow**
The auth system uses JWT tokens with refresh rotation. Login requests hit
`src/auth/login.ts` which validates credentials against the user store,
then issues a token pair via `src/auth/tokens.ts`.
-> Wiki: "Authentication Architecture" · "Token Lifecycle"
**Related files:** src/auth/login.ts, src/auth/tokens.ts, src/middleware/auth.ts
If no wiki exists for this repo, suggest the user set one up in the cubic dashboard.