CortexGraph FAQs

Question 1

Is my data private and secure with CortexGraph?

Accepted Answer

Yes, privacy is a core feature. All data is stored locally on your machine in human-readable JSONL for short-term and Obsidian-compatible Markdown for long-term memory. No cloud services are involved, ensuring complete control and transparency over your data.

Question 2

How does CortexGraph integrate with AI assistants or LLMs?

Accepted Answer

CortexGraph uses a smart prompting system for automatic memory saving, recall, and reinforcement directly within conversations. This allows AI assistants to naturally leverage their memory without requiring explicit memory commands from the user.

Question 3

What kind of memory storage does CortexGraph use?

Accepted Answer

It features a two-layer memory architecture: short-term memory is stored in local, Git-friendly JSONL files, while long-term, permanent memories are saved as Obsidian-compatible Markdown files, complete with YAML frontmatter and Wikilinks.

Question 4

What is CortexGraph and what problem does it solve?

Accepted Answer

CortexGraph provides AI assistants with a human-like temporal memory system. It solves the common problem of AI forgetting past conversations by enabling natural memory decay and reinforcement, making interactions more continuous and personalized.

Question 5

How does CortexGraph mimic human-like memory dynamics?

Accepted Answer

It implements a novel temporal decay algorithm based on the Ebbinghaus forgetting curve, combined with reinforcement learning. Memories naturally fade unless reinforced through use, and frequently used information is promoted to long-term storage.

CortexGraph

CortexGraph

주요 기능

사용 사례

주요 기능

사용 사례