Reflect Memory FAQs

Question 1

How can developers integrate Reflect Memory into their applications?

Accepted Answer

Developers can integrate Reflect Memory flexibly using its TypeScript SDK, the MCP (Memory Control Plane) Server, or directly via its comprehensive REST API, making it suitable for various development workflows.

Question 2

What is Reflect Memory?

Accepted Answer

Reflect Memory is a privacy-first, vendor-neutral memory layer designed for AI agents. It provides a shared, persistent, and user-controlled context, allowing AI models to access structured, editable, and deletable information only when you explicitly permit it.

Question 3

Which AI agents and LLMs does Reflect Memory support?

Accepted Answer

Reflect Memory offers vendor-neutral support for major AI agents, including ChatGPT, Claude, Gemini, and Grok. It's designed to integrate with any OpenAI-compatible API key or local models like Ollama.

Question 4

How does Reflect Memory ensure privacy and user control?

Accepted Answer

It implements explicit user control, meaning all memory is authored, structured, editable, and deletable by the user. AI models are stateless and only see what you choose to show them, with deterministic visibility rules managed by `allowed_vendors`.

Question 5

What is agent-scoped memory access?

Accepted Answer

Agent-scoped memory access ensures that AI agents only see specific memories they are authorized to access. Memories can be tagged with `allowed_vendors`, restricting visibility to designated agents (e.g., only ChatGPT) or making them universally visible to all agents with `"*"`.

Reflect Memory

Reflect Memory

주요 기능

사용 사례

주요 기능

사용 사례