bcharleson

OpenShart — encrypted AI agent memory with OpenClaw plugin support for secure memory_search/get/store/forget workflows.

15
5
89% credibility
Found Feb 19, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

OpenShart provides encrypted, fragmented storage for AI agent memories with searchable encryption, access controls, and compliance features for secure data handling.

How It Works

1
🔍 Discover Secure AI Memory

You hear about a clever way to keep your AI helper's thoughts safe and private, like a super-secure diary.

2
📦 Set Up Your Vault

You grab the tool and pick a simple place to store memories, like right in your app or a small file.

3
🔑 Create Your Secret Key

You make a strong private password that locks everything away, just for you.

4
💾 Save Your First Memory

You tell your AI to remember something important, and it automatically spots sensitive info and splits it into safe pieces.

5
🔒 Search Without Worry

You ask your AI to find old thoughts, and it pulls them back perfectly without ever risking the full info.

6
🗑️ Erase When Done

You decide to forget something forever, and it wipes it clean so no trace remains.

AI Remembers Safely

Your AI now has a rock-solid memory that stays private, compliant, and ready for anything, giving you total peace of mind.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 15 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 openshart?

OpenShart is a TypeScript library for encrypted AI agent memory, splitting data across Shamir-shared fragments with AES-256-GCM so no single breach exposes everything. It handles secure memory_search, get, store, and forget workflows via a simple API, plus OpenClaw plugin support for plug-and-play integration. Devs get enterprise-grade persistence—FIPS-ready, searchable without decryption, with backends like Postgres or SQLite—solving agent memory leaks in compliant apps.

Why is it gaining traction?

Zero runtime deps, presets like "government" for ChainLock timing locks and Bell-LaPadula access, stand out versus basic encrypted stores. Searchable encryption via HMAC tokens and auto-PII detection with redaction hook privacy-focused devs, while HIPAA/SOC2/GDPR designs ease audits. OpenShart online demos and OpenClaw plugin make agent workflows secure without custom crypto plumbing.

Who should use this?

AI agent builders on OpenClaw or LangChain needing tamper-proof memory for production. Compliance engineers handling PHI or classified data in TypeScript backends. Teams evaluating openshare github for secure, distributed agent storage without vendor lock-in.

Verdict

Solid foundation with 64 passing tests and distributed validation scripts, but 10 stars and 0.90% credibility score signal early maturity—docs are thorough, though real-world scale unproven. Try for prototypes if security trumps polish; skip for mission-critical without audits.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.