AI Knowledge System

by kellyoconor

Not rated
GitHub

About

An enterprise-ready system to archive AI conversations from ChatGPT and Claude into a Supabase database.

Details

Author
kellyoconor
Categories
Database, Other, AI, Knowledge Base

Setup

Install AI Knowledge System in your MCP client (Claude Desktop, Cursor, Windsurf, and others).

Repository: https://github.com/kellyoconor/chatgpt-supabase-api

Follow the installation instructions in the repository README, then restart your MCP client.

Enterprise-Ready Conversation Archive for ChatGPT + Claude

Transform your AI conversations from lost chat history into a searchable, organized knowledge base. This system captures insights from both ChatGPT Enterprise and Claude Desktop, storing everything in a structured database for long-term reference and analysis.

- πŸ”„Dual AI Integration: Seamlessly capture conversations from both ChatGPT and Claude
- 🏒Enterprise Compliant: Works with ChatGPT Enterprise restrictions via browser extension
- πŸ“ŠSmart Organization: Auto-categorize by strategy, user feedback, competitive analysis, etc.
- 🎯Priority Levels: Mark conversations as low, medium, high, or critical importance
- πŸ“…Time-Based Filtering: Automatic quarterly tagging for periodic reviews
- πŸ”Powerful Search: Query across months of conversations to find patterns and insights
- πŸ›‘οΈPrivacy First: Your data stays in your own Supabase database
- ⚑One-Click Save: Browser extension makes saving ChatGPT conversations effortless

- Product Managersbuilding competitive intelligence and user research repositories
- Consultantsmaintaining client insight databases
- Researchersorganizing AI-assisted analysis sessions
- Teamscreating shared knowledge bases from AI conversations
- Anyonewho wants to stop losing valuable AI insights

- Node.js 18+
- Chrome browser
- Supabase account (free tier works)

-- Create your conversations table CREATE TABLE conversations ( id SERIAL PRIMARY KEY, topic VARCHAR(100), content TEXT NOT NULL, category VARCHAR(50), priority VARCHAR(10), source VARCHAR(20), tags TEXT[], conversation_date DATE DEFAULT CURRENT_DATE, quarter VARCHAR(10), created_at TIMESTAMP DEFAULT NOW() );
git clone https://github.com/yourusername/ai-knowledge-system cd ai-knowledge-system/api-server npm install npm start

- Openchrome://extensions/
- Enable "Developer mode"
- Click "Load unpacked"
- Select thebrowser-extensionfolder

Add to~/.claude/claude_desktop_config.json:

{ "mcpServers": { "supabase": { "command": "node", "args": ["/path/to/supabase-mcp/packages/mcp-server-supabase/dist/transports/stdio.js"], "env": { "SUPABASE_ACCESS_TOKEN": "your_token", "SUPABASE_URL": "your_url", "SUPABASE_ANON_KEY": "your_key", "SUPABASE_PROJECT_REF": "your_project" } } } }

- Have your conversation normally
- Click the extension icon
- Review auto-suggested topic
- Select category and priority
- Click "Save Conversation"

"Save this strategy conversation with high priority: [your conversation content]" "Show me all user feedback insights from Q1 2025" "Find high-priority competitive analysis discussions"
ChatGPT Enterprise β†’ Browser Extension β†’ API Server β†’ Supabase Claude Desktop β†’ MCP Protocol β†’ Local Server β†’ Supabase

- Browser Extension: Captures ChatGPT conversations with enterprise compliance
- API Server: Hosted endpoint for saving conversations to database
- MCP Integration: Direct Claude Desktop to database connection
- Supabase Database: PostgreSQL with structured conversation storage

ai-knowledge-system/ β”œβ”€β”€ browser-extension/ β”‚ β”œβ”€β”€ manifest.json β”‚ β”œβ”€β”€ popup.html β”‚ β”œβ”€β”€ popup.js β”‚ └── content.js β”œβ”€β”€ api-server/ β”‚ β”œβ”€β”€ server.js β”‚ β”œβ”€β”€ package.json β”‚ └── .env.example β”œβ”€β”€ mcp-setup/ β”‚ └── claude_desktop_config.json β”œβ”€β”€ database/ β”‚ └── schema.sql └── docs/ └── technical-overview.html

- strategy- High-level planning and direction
- user_feedback- Customer insights and research
- competitive_analysis- Market and competitor research
- brainstorming- Creative ideation sessions
- general- Miscellaneous conversations

- critical- Mission-critical insights
- high- Important for near-term decisions
- medium- Useful reference material
- low- Nice-to-have information

# API Server cd api-server npm install npm run dev # Extension Development # Load unpacked extension in Chrome for hot reloading

- New Categories: Update database constraints and extension dropdown
- Custom Fields: Extend database schema and forms
- Integrations: Add new AI tools following the extension pattern
- Fork the repository
- Create a feature branch (git checkout -b feature/amazing-feature)
- Commit your changes (git commit -m 'Add amazing feature')
- Push to the branch (git push origin feature/amazing-feature)
- Open a Pull Request

- Slack integration for team sharing
- Export to PDF/Word reports
- Advanced analytics dashboard
- Mobile app for conversation capture
- Integration with Notion/Obsidian
- Multi-user support with permissions

This system is designed to work within enterprise constraints:

- βœ… No API keys required for ChatGPT Enterprise
- βœ… Data stays in your own database
- βœ… Browser extension uses standard web APIs
- βœ… No external AI service dependencies

MIT License - seeLICENSEfile for details.

- Supabase MCP Communityfor the foundational MCP server
-
Anthropicfor Claude and MCP protocol
-
OpenAIfor ChatGPT platform

- Create anissuefor bugs
- Start a
discussionfor questions
- Follow
@yourusernamefor updates

Built by product managers, for product managers.πŸš€

Stop losing your best AI insights. Start building your knowledge advantage.

Engram is a hosted MCP server that provides reliable memory for AI agents:

Persistent memory, teams, and projects for AI agents. 76 MCP tools for storing, recalling, and sharing knowledge across sessions.

A place where AI Agents can ask and answer questions and share knowledge

Deterministic O(1) memory for AI agents β€” local-first, MCP-native, with multi-agent speaker attribution and millisecond recall.

A free, plug-and-play knowledge base with over 10,000 built-in insight reports and support for parsing private documents.

A memory management system for AI assistants to store, retrieve, and manage user information using a local database.

A runtime-configurable MCP server that turns a Supabase project into an AI-compatible tool interface.

Let AI read, search, and summarize your UpNote notes β€” and create new ones β€” right from your chat.

Agent memory with git-like version control. Custom LLMs turn conversations into structured facts with automatic conflict detection - your agent sees how decisions evolved, not four contradictory text blobs. 80% token reduction vs RAG/graph systems. MCP + REST.

Production-ready RAG out of the box to search and retrieve data from your own documents.

No reviews yet β€” be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.