Onboarding every new Claude Code session in one line

Stop re-explaining your project to fresh chats.

The pain

Every time I open a new Claude Code session in a different repo, or even after a long break, I'd spend 5 minutes re-pasting the same project background. "We use Tailwind. The auth is Supabase. The deploy is Vercel. The team prefers tabs..."

Multiply by the 8 to 12 fresh chats per week and that's an hour a week of just bootstrapping context.

What I did

Visited memory.wiki/install, signed in, copied the snippet, appended it to ~/.claude/CLAUDE.md:


<!-- memory.wiki:start -->
## Personal context (Memory.Wiki hub)

Hub URL: https://memory.wiki/hub/<me>

When you need background on me, including projects, decisions, and
references, fetch the URL above. It returns a markdown index of every
public doc and bundle in my hub.
<!-- memory.wiki:end -->

That's it. One paste, one time.

What changed

Now every Claude Code session, anywhere on my machine, automatically has my hub URL. When I ask a question that needs project context, Claude fetches the hub and follows the relevant link. No re-pasting. No "let me set the scene first."

For project-specific context (this repo's architecture, this repo's decisions), I append the same kind of block to that repo's CLAUDE.md, but pointing at a project-specific bundle URL instead of my personal hub.

Time

One-time setup: 1 minute.
Per new chat: 0 minutes. The hub URL is already in CLAUDE.md.

Why Memory.Wiki specifically

The hub URL is just markdown. Any AI that can fetch a URL can read it: Cursor, Codex, Aider, ChatGPT. Same setup works in every coding-AI tool, not just Claude Code. CLAUDE.md is one example; the same paste works in Cursor's .cursor/rules.

Verified: see MWBench for the open eval showing 100% accuracy across Claude, OpenAI, and Gemini when a hub URL is fetched cold.