Skip to content
View robertogogoni's full-sized avatar
💭
🧠
💭
🧠
  • São Paulo, Brazil
  • 05:40 (UTC -03:00)

Block or report robertogogoni

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
robertogogoni/README.md

Typing SVG

Scrum Master who codes on the side. If something's broken or tedious, I'll probably write a script for it.

Sao Paulo, Brazil

LinkedIn X


Tech Stack

Node.js JavaScript SQLite Bash PowerShell Linux Windows Arch Linux Git GitHub Actions


Featured Project

Claude's Cognitive Layer - A dual-model memory system with auto-extraction, auto-recall, vector search, MCP tools for deep reasoning, and compounding learnings.

                         +---------------------------+
                         |    Claude Code Session    |
                         +-------------+-------------+
                                       |
            +==========================|==========================+
            |                   CORTEX MCP SERVER                 |
            +=====================================================+
            |                                                     |
            |   +------------------+     +-------------------+    |
            |   |   HAIKU WORKER   |     |  SONNET THINKER   |    |
            |   |                  |     |                   |    |
            |   |  o query         |     |  * reflect        |    |
            |   |  o recall        |     |  * infer          |    |
            |   |                  |     |  * learn          |    |
            |   |  ~$0.25/1M [fast]|     |  * consolidate    |    |
            |   +--------+---------+     |  ~$3/1M tokens    |    |
            |            |               +--------+----------+    |
            |            +--------+------+--------+               |
            |                     |                               |
            |   +-----------------v-----------------------+       |
            |   |        VECTOR SEARCH ENGINE             |       |
            |   |   HNSW Index + BM25 + RRF Fusion        |       |
            |   |   Local Embeddings (all-MiniLM-L6)      |       |
            |   +-----------------------------------------+       |
            |                                                     |
            +=====================================================+
                          |                       |
            +-------------+----------+  +---------+-------------+
            |     SESSION START      |  |      SESSION END      |
            |------------------------|  |-----------------------|
            |  - Context Analyzer    |  |  - Extraction Engine  |
            |  - Query Orchestrator  |  |  - Pattern Tracker    |
            |  - Memory Injection    |  |  - Outcome Scorer     |
            +------------------------+  +-----------------------+
                          |                       |
                          +-----------+-----------+
                                      |
            +-------------------------v-------------------------+
            |               LADS SELF-IMPROVEMENT               |
            |     Learnable | Adaptive | Documenting | Self-I   |
            +---------------------------------------------------+
Problem Cortex Solution
Claude forgets context Auto-recall: Injects relevant memories at session start
Learnings are lost Auto-extraction: Captures insights from every session
No semantic search Vector Search: HNSW + BM25 hybrid with RRF fusion
No deep reasoning tools MCP Server: 6 tools for query, recall, reflect, infer, learn, consolidate
Expensive API calls Dual-model: Haiku for fast ops, Sonnet for deep reasoning
Manual memory management Fully automatic: Zero user intervention required

NEW: /cortex Skill - User-friendly slash commands:

/cortex              # Status overview
/cortex help         # All commands with examples
/cortex query "X"    # Search memories (~$0.001)
/cortex learn "X"    # Store insight (~$0.01)
/cortex reflect "X"  # Deep analysis (~$0.01)
/cortex stats        # Memory counts & API costs
/cortex health       # System health check
/cortex export       # Export memories (json/md)

Tests Node.js MCP Vector Search License: MIT Roadmap

# One-command install
git clone https://github.com/robertogogoni/cortex-claude.git ~/.claude/memory
cd ~/.claude/memory && npm install

Other Projects

claude-cross-machine-sync update-beeper

claude-cross-machine-sync - AI-powered config sync across Windows & Linux with intelligent auto-categorization and searchable AI memory.

update-beeper - Self-healing Beeper Desktop updater for Arch Linux with automatic recovery and rollback.

Lint License: MIT


GitHub Stats

Profile Views

Trophies

GitHub Trophies

Overview

GitHub Stats Top Languages

Streak

GitHub Streak

Activity

Activity Graph

Contributions

github contribution grid snake animation

Summary Cards

Profile Details

Repos per Language Most Commit Language

Stats Card Productive Time


Notes & Documentation

Personal troubleshooting guides and learnings.

Topic Description
Claude in Chrome Troubleshooting Fixes for extension detection issues, native messaging setup, and bug #19911

Building practical solutions for everyday problems

Pinned Loading

  1. beeper-community/update-beeper beeper-community/update-beeper Public

    Self-healing Beeper Desktop updater for Linux with automatic rollback

    Shell 1