rizal72

rizal72 / true-mem

Public

A persistent memory plugin for OpenCode with cognitive psychology-based memory management

24
0
100% credibility
Found Feb 28, 2026 at 21 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

True-Mem is a plugin for the OpenCode AI coding assistant that automatically captures, stores, and recalls user preferences, project decisions, and key learnings from conversations across sessions.

How It Works

1
💡 Discover True-Mem

You hear about True-Mem, a smart helper that makes your AI coding buddy remember your favorite styles and choices forever, so no more repeating yourself.

2
âž• Add it to your AI buddy

Simply tell your AI coding assistant to include True-Mem in its list of trusted helpers, like picking a new favorite tool.

3
🔔 Welcome message appears

Restart your assistant and smile at the cheerful popup saying 'Memory active' – everything is set up and ready to go!

4
💬 Chat naturally

Talk to your AI as usual, sharing things like 'I prefer clean code' or 'Use this approach for projects' – it listens quietly in the background.

5
🧠 It remembers automatically

Open a new chat or project later, and your AI brings up your past choices right away, feeling like an old friend who knows you well.

6
📋 Check or tweak memories

Say 'show memories' to see the list, or ask it to forget something specific if needed – total control at your fingertips.

🎉 Never repeat yourself again

Your coding sessions flow smoothly with a personalized AI that recalls your style and decisions perfectly every time.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 21 to 24 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 true-mem?

True-mem is a TypeScript plugin for OpenCode that delivers persistent memory across AI coding sessions, capturing prefs like "prefer TypeScript over JS" or decisions like "use SQLite here" from chats and injecting them into prompts. It solves AI forgetfulness with github persistent storage via built-in SQLite, offering global or project scopes without repeating yourself. Modeled on cognitive psych—Ebbinghaus decay for episodic facts, permanent for constraints—it filters noise like questions or AI meta-talk.

Why is it gaining traction?

Unlike basic keyword loggers, it uses psych-inspired scoring (recency, utility, etc.) and four-layer defenses to slash false positives, plus multilingual patterns in 15 languages for real-world chats. Non-blocking async extraction keeps OpenCode snappy, with user commands like "list-memories" or "delete true-mem memory about bun" for control. For persistent memory llm setups, it's zero-dep (Bun/Node), beating ram vs persistent memory limits without heavy vector stores.

Who should use this?

OpenCode power users tired of reteaching AI "never use var" or project choices like REST over GraphQL. Solo devs needing github persistent identifier for cross-project styles, or teams enforcing "always run tests first" via global memories. Pairs well with persistent memory pmem workflows if you're scaling AI-assisted coding.

Verdict

Grab it if you're deep in OpenCode—installs via config tweak, toasts confirm load—but 19 stars and 1.0% credibility scream early alpha. Great docs and debug tools make testing low-risk; fork and contribute to mature this persistent memory module.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.