LWC
Persistent, source-grounded memory system for AI agents across sessions
Visit janyork.github.io ↗What is LWC?
LWC is a CLI tool that gives AI agents durable memory for knowledge, decisions, and plans that persists across work sessions. It stores information locally as searchable wiki pages with full source attribution and change history, letting agents pick up context where previous work left off. The system works with multiple AI coding assistants including Claude Code, Cursor, and GitHub Copilot.
Designed for both individual and team use, LWC emphasizes local ownership and control. Teams can share knowledge spaces with permission-based access, host the system on their own infrastructure, and maintain an auditable record of memory changes. The tool focuses on preserving sources and connections rather than embedding its own language model, allowing agents to reason over curated, traceable information.
Momentum
Daily activity over the last 30 days: upvotes, unique visits, new GitHub stars and Hacker News points.
How do we know it's vibe coded?
vibetrend discovered LWC automatically. Evidence: GitHub JanYork/llm-wiki-cli; repo has AGENTS.md. View source.
Similar vibe-coded apps

Self-hosted AI agent with persistent memory and multi-platform messaging

Community marketplace for DeepSeek Harness Web themes, pets, plugins

Open-source, AI-native knowledge base with Evernote-style interface

Ultra-lightweight, self-hosted open-source AI agent framework

Web and mobile interface for Hermes Agent with cross-device session sync

Open-source desktop AI agent app for macOS, Windows, and Linux

Open source AI operating system for company productivity and agent workflows
