v0.2.0-beta · Apache 2.0 · PostgreSQL Native

Universal Knowledge Infrastructure

OpenCortyx is an open knowledge layer where anything can become connected knowledge, and anyone or any AI agent can retrieve the context they need from it.
Request early access See how it works
Self-Hostable PostgreSQL Native Model Context Protocol Zero Lock-In
How It Works

Knowledge is everywhere. Context shouldn't be.

Information lives scattered across tables, notes, code, and chat threads. OpenCortyx brings them together into one unified layer so you can query what matters.

Your World
Database
Documents
Code
Memory
Conversations
APIs
Events
Anything else
OpenCortyx
Knowledge Layer
Local or Cloud
PostgreSQL Native
Model Context Protocol
Retrieved Context
Human (Developers, Teams, Researchers)
AI Tools (Cursor, Claude, Coding Assistants)
Applications (Internal Tools, Production Systems)
Beyond Chat Logs

Memory is just one kind of knowledge.

Most AI tools focus only on temporary chat transcripts. But real understanding requires connecting every type of knowledge your team relies on.

A person might bring:
  • Documents
  • Notes
  • Research
  • Personal knowledge
  • Domain insights
A system might bring:
  • Databases
  • Schemas
  • APIs
  • Events
  • Telemetry
An agent might bring:
  • Memories
  • Observations
  • Conversations
  • Decisions
  • Experiences
OpenCortyx connects all of them into a single, retrievable context layer.
Connect Anything

Knowledge has no fixed shape.

Anything can become knowledge. Everything can be connected. The world does not come in one schema. Knowledge can be a table, a document, a memory, a person, an event, a claim, an observation, a relationship, or something entirely new. OpenCortyx lets you define what matters, connect it, and make it retrievable.

Person works at ➔ Company
Person knows ➔ Concept
Document supports ➔ Claim
Claim contradicts ➔ Claim
Memory about ➔ Person
Transaction belongs to ➔ Account
Agent observed ➔ Event
Anything relates to ➔ Anything
Context on Demand

Get the exact context you need. Without the noise.

When you or an AI assistant ask a question, OpenCortyx traces the connections across your notes, code, and team history to retrieve the exact context you need.

Question "What changed in this system, why did it change, and what does it affect?"
How OpenCortyx connects the dots across your systems
01 Question Incoming query Identifies core subject
02 Concept Auth Flow Connects domain logic
03 System Billing Engine Locates active service
04 Schema Users Table Finds modified schema
05 Change Migration #104 Added 2FA enforcement
06 Decision RFC 42 Approved Security compliance rule
07 Conversation Team Review Edge case discussion
08 Evidence Commit 4f98a12 Verified source of record
The Answer Delivered to You
A clear, verified answer backed by real team history, documentation, and code commits, without guesswork.
Connected Context Ready for You or Your Tools 412 tokens · Relevant facts only 
{
  "query": "What changed in this system, why did it change, and what does it affect?",
  "grounded_facts": [
    {
      "entity": "Billing Service (v2.1)",
      "change": "Enforced Two-Factor Authentication for credential updates",
      "schema_migration": "db/migrations/104_users_mfa_column.sql",
      "author": "Sarah Lin (Engineering Lead)",
      "git_commit": "4f98a12 (verified)"
    }
  ],
  "decision_provenance": {
    "document": "RFC 42: Mandatory 2FA for Financial Actions",
    "approved_by": "Security Review Board on 2026-08-14",
    "discussion_ref": "#sec-ops thread 2026-08-12: Handling legacy API keys",
    "source_citation": "docs/rfcs/rfc-0042.md:L45-L72"
  },
  "upstream_impact": [
    "Mobile checkout client requires SDK v3.4 update",
    "Legacy automated webhook bypass deprecated"
  ],
  "status": "verified"
}
One Source of Truth

Built for people. Ready for every tool and agent.

Instead of repeating yourself across disconnected apps while your AI assistants forget everything between sessions, OpenCortyx gives your whole workflow one shared, persistent source of truth.

You & Your Team (Developers, Researchers)
Your AI Tools (Cursor, Claude, Copilot)
Your Systems (APIs, Codebases, Workflows)
OpenCortyx Connected Knowledge Layer You and every tool you use share the exact same connected context through open protocols like the Model Context Protocol (MCP).
Frequently Asked Questions

Common questions about OpenCortyx.

Q: What is OpenCortyx in simple terms?
OpenCortyx is a single place where you can bring together data, documents, memories, facts, and relationships. Instead of searching across disconnected tools or re-explaining projects to an AI every day, you ask a question and get the connected context behind it.
Q: Why not just use a vector database?
Vector search only looks for words that seem mathematically similar. It does not know who approved a decision, which database table is affected by a document, or whether two claims contradict each other. OpenCortyx connects the relationships between facts, not just matching words.
Q: How do AI agents and tools connect to OpenCortyx?
Through standard interfaces like the Model Context Protocol (MCP). You connect your agent, IDE, or script in a few lines of configuration, and it can immediately query the shared knowledge layer.
Q: Where does my data live?
On your own infrastructure. You can run OpenCortyx locally on your machine with PostgreSQL. Your notes, schemas, and conversations stay private and under your complete control.
Q: Why is OpenCortyx in private beta?
We are actively refining OpenCortyx with a select group of developers, founders, and researchers before open release. Request access below to join the developer beta.

Build your knowledge layer.

Bring your data, documents, and team knowledge together. Request access to the private developer beta.

OpenCortyx is in private developer beta. Public release and community coming soon.