Smriti FAQs

Question 1

What is Smriti?

Accepted Answer

Smriti is a standalone, local-first MCP server that provides persistent and semantically searchable memory for AI agents. It operates entirely without cloud dependencies, ensuring privacy and offline functionality.

Question 2

How does Smriti provide semantic search capabilities?

Accepted Answer

Smriti utilizes `sqlite-vec` and a local embedding model (like all-MiniLM-L6-v2 via ONNX) to generate vector embeddings for your captured memories. This enables semantic search, allowing you to find information based on meaning rather than just keywords.

Question 3

How does Smriti ensure data privacy?

Accepted Answer

Smriti is designed for 100% local operation. All data, including vector embeddings and memories, stays on your machine in a single SQLite file. There are no cloud dependencies, external servers, or API keys, guaranteeing complete privacy.

Question 4

What are the key benefits of using Smriti over cloud-based memory solutions?

Accepted Answer

Smriti offers significant advantages including zero cost, full offline functionality, superior privacy (all data stays local), easy portability (single .db file), and a simpler setup process compared to cloud alternatives that often require accounts, API keys, and variable pricing.

Question 5

What kind of AI agents can use Smriti?

Accepted Answer

Smriti is compatible with any MCP (Memory, Context, and Prompt) client. This includes popular tools like Claude Code and Cursor, allowing multiple agents to share a unified, persistent memory store.

Smriti

Smriti

主要功能

使用案例

主要功能

使用案例