cold-start

Import and sequence data sources into linked wiki pages.

5|1|Updated Apr 15, 2026
One-click install
npx skills add https://github.com/GYF0311/lorekit --skill cold-start-gyf0311
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cold-start
Source: https://github.com/GYF0311/lorekit/tree/main/brain/skills/cold-start
Command: npx skills add https://github.com/GYF0311/lorekit --skill cold-start-gyf0311

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the “empty brain” problem by turning day-one chaos—email, contacts, calendar events, conversations, social archives, and file history—into a linked, searchable knowledge base quickly and safely.

Core Features & Use Cases

  • Day-one phased data import: Sequences the highest-leverage sources so each phase compounds the next (markdown/Obsidian first, then people, calendar, Gmail, conversations, X/Twitter, archives, and transcripts).
  • Safety-first credential handling via ClawVisor: Prevents the agent from ever receiving raw OAuth tokens by requiring a credential gateway and user consent for each import phase.
  • Entity detection and cross-linking during ingestion: Detects and updates people/company pages while importing, links attendees to meetings, and enriches context as it goes.
  • Resumable progress tracking: Records phase completion in ~/.gbrain/cold-start-state.json so interrupted sessions resume without repeating work.

Quick Start

Ask the AI to run a cold start import for your brain and walk you through each phase with consent, using ClawVisor for Gmail/Calendar/Contacts if you enable it.

Frequently Asked Questions about cold-start

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I import emails, contacts, and calendar events into a knowledge base?▼

To import emails, contacts, and calendar events into a knowledge base, this skill performs a phased day-one ingestion using ClawVisor credential mediation to safely connect Google APIs without exposing raw OAuth tokens.

What is the best way to migrate Obsidian markdown files into a searchable wiki?▼

Migrating Obsidian markdown files into a searchable wiki is handled by sequencing markdown migration as the first high-leverage import phase, establishing persistent cross-linked wiki pages before enriching them with other data sources.

Can I import a Twitter archive and meeting transcripts into my personal knowledge base?▼

Yes, you can import a Twitter archive and meeting transcripts into your personal knowledge base; the ingestion process detects entities, updates timeline links, and compiles searchable context from conversation exports and file archives.

How does entity resolution work during day-one data ingestion?▼

Entity resolution during day-one data ingestion works by automatically detecting people and companies during import, linking meeting attendees to events, and updating cross-links across persistent wiki pages as each phase compounds.

What happens if an import session is interrupted while bootstrapping my knowledge base?▼

If an import session is interrupted while bootstrapping your knowledge base, resumable progress tracking saves phase completion in a local state file, allowing the ingestion to resume without repeating previously completed work.

Do I need to provide raw OAuth tokens to sync Gmail and Calendar data?▼

No, you do not need to provide raw OAuth tokens to sync Gmail and Calendar data; the ClawVisor credential gateway acts as a consent-gated mediator, preventing the agent from ever receiving raw credentials during import.