brain

Manage a SQLite-backed project knowledge base for structured memory.

264|11|Updated Apr 25, 2026
One-click install
npx skills add https://github.com/coco-research/coco --skill brain-coco-research
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: brain
Source: https://github.com/coco-research/coco/tree/main/systems/brain/skills/brain
Command: npx skills add https://github.com/coco-research/coco --skill brain-coco-research

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires chromadb, docx, PyPDF2, openpyxl, pptx, and includes scripts (resource) components.

What problem does it solve?

This Skill replaces scattered notes and unreliable chat memory with a structured SQLite knowledge base for each project folder, making project context durable, searchable, and easy to update.

Core Features & Use Cases

  • Project memory management: Stores entities, relationships, tasks, threads, decisions, and events in project_brain.db.
  • Session context and search: Generates concise project briefings, open-task views, recent decisions, and full-text style searches for active work.
  • Sync and export workflows: Supports deterministic CLI operations, folder scanning, brain.json sync, MemPalace syncing, and CLAUDE.local.md export for longer-running projects.
  • Use Case: A developer can capture a new stakeholder, link them to a team, record a decision, and then retrieve the current project state in one consistent knowledge system.

Quick Start

Ask the brain skill to initialize or query the current project folder so it can record open tasks, recent decisions, and key entities in project_brain.db.

Frequently Asked Questions about brain

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

FAQPage Schema
How do I create a persistent project knowledge base for tracking decisions and tasks?

To create a persistent project knowledge base, initialize the skill in your workspace folder to generate a SQLite database. This database captures entities, relationships, tasks, threads, decisions, and events, replacing scattered notes with structured, searchable project memory.

Can I use SQLite for deterministic session context and project memory management?

Yes, SQLite enables deterministic database operations and idempotent updates for reliable project memory management. It structures session context and relational queries, ensuring your project history remains consistent across repeated development work.

What's the best way to sync project history to an external memory store?

The best way to sync project history is using the built-in brain.json synchronization and MemPalace syncing features. These tools export your structured project state to external memory stores, ensuring durable context for longer-running projects.

Does this project memory tool work with codebases and workspace folders?

Yes, this tool applies directly to codebases and workspace folders that need structured memory. It scans folders to record open tasks, recent decisions, and key entities, providing searchable project history across active development work.

How do I export project memory and session context for longer-running projects?

You can export project memory and session context by generating a CLAUDE.local.md file. This feature creates concise project briefings, open-task views, and recent decisions, making structured project state accessible for longer-running workflows.

Why use a SQLite database for decision logs and task tracking instead of chat memory?

A SQLite database provides reliable, relational queries and full-text style searches that unreliable chat memory cannot. It ensures deterministic updates and durable project state, capturing stakeholders, teams, and decisions in one consistent knowledge system.