catello09

catello09 / agent-memory-store

Public

TypeScript agent memory layer: semantic vector recall + SQLite-backed storage, Chroma or in-memory vectors, REST API, MIT.

agent-memory ai-agents chromadb embeddings hybrid-search
20
0
100% credibility
Found Apr 18, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A TypeScript library that provides persistent storage for AI agent memories using semantic similarity search combined with structured, filterable records.

How It Works

1
🧠 Discover memory for AI helpers

You want your AI assistant to remember conversations, preferences, and facts forever instead of forgetting everything.

2
📥 Bring home the memory keeper

Grab this simple tool designed to give your AI a persistent brain that saves details on your computer.

3
🔗 Link a thinking service

Connect a smart service so it can understand and match similar ideas automatically, or use a quick stand-in for testing.

4
💾 Save special memories

Tell it to store notes like 'user loves bullet points' with tags for type, time, and extra details.

5
🔍 Find matching memories

Ask about a topic and get back the most similar past notes, or filter by type, date, or importance.

6
🌐 Share via simple web access

Turn on a friendly web door to add, search, or tidy up memories from anywhere.

🎉 AI never forgets again

Your helper now recalls everything perfectly across sessions, making chats smarter and more personal.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is agent-memory-store?

This TypeScript agent memory store gives AI agents persistent recall across sessions, blending semantic vector search for fuzzy matches with SQLite-backed rows for structured queries on metadata, timestamps, and types like episodic, semantic, or procedural. Developers get a simple API to remember content, recall similar items via queries, filter with SQL-like wheres, and serve it all over a REST API on port 3000. It supports OpenAI, Ollama, or hash embedders, with Chroma for production vectors or in-memory for tests, all configurable via env vars.

Why is it gaining traction?

It stands out in the TypeScript agent AI framework space by packaging hybrid storage—vectors plus transactional rows—into one lightweight layer, dodging the setup pain of separate vector DBs and SQL. The REST endpoints for recall, query, consolidate, and snapshots make it drop-in ready for agentic AI prototypes, with Docker Compose bundling Chroma and the API. Pluggable consolidation rules auto-merge duplicates, a hook for scaling TypeScript agents without manual cleanup.

Who should use this?

Backend devs building TypeScript agent libraries or skills for chatbots, like those integrating Claude or OpenAI in Node.js apps needing context retention. It's ideal for TypeScript agentic frameworks where agents must query past interactions by similarity or filters, such as customer support bots tracking user preferences or multi-turn reasoning agents.

Verdict

Promising agent memory store for TypeScript AI experiments—grab it if you need quick persistent recall without infrastructure headaches, backed by solid docs, tests, and MIT license. At 20 stars and 1.0% credibility, it's early-stage raw; production users should add monitoring before committing.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.