43 lines (36 loc) · 1.33 KB

AI Trinity Data Extraction Project

Objective

Set up Claude Code and Codex with all MCPs, then extract and map data from:

Cloze CRM (handled by Claude Code)
Notion (handled by Codex o3)

Challenge & Solution

Claude Code: Has native MCP support - can directly access Cloze
Codex: No native MCP support - will use Claude Code as a bridge to access Notion

MCP Servers Available

Based on our system check, these MCPs are configured:

mem0 - Shared memory
notion - Notion workspace access
cloze - CRM data access
filesystem - File operations
desktop-commander - System commands
github - Repository access
supabase - Database operations
figma - Design files
cognitive-os-obsidian - Notes
cognitive-os-whoop - Health data
cognitive-os-activitywatch - Activity tracking
playwright - Browser automation
sequential-thinking - Complex reasoning
web-search - Brave search API

Execution Plan

Phase 1: MCP Configuration

Ensure Claude Code has access to all MCPs
Create data bridge for Codex

Phase 2: Data Extraction

Claude Code extracts Cloze CRM data
Claude Code accesses Notion data and saves for Codex
Codex analyzes and maps the Notion data

Phase 3: Data Synthesis

Combine insights from both sources
Create unified data map
Generate actionable insights