mnlt/wellread

Overview

Wellread is a TypeScript library that gives AI agents a shared, persistent memory for research context. It stores and retrieves structured knowledge across sessions, enabling agents to build on past findings without repeating work.

Best for

Best for
Developers building multi-agent systems that need persistent research memory

Use cases

• Persisting research context across multiple agent conversations
• Building a shared knowledge base for collaborative AI agents
• Reducing redundant queries by caching prior research results

How to use

Install

npx wellread

Tested with

Claude Code, Cursor, Windsurf, Gemini CLI, VS Code, OpenCode

Notes

Wellread is a TypeScript library that gives AI agents a shared, persistent memory for research context. It stores and retrieves structured knowledge across sessions, enabling agents to build on past findings without repeating work.

8 stars on GitHub. Last updated 2026-04-19. Licensed AGPL-3.0.

Use cases

• Persisting research context across multiple agent conversations
• Building a shared knowledge base for collaborative AI agents
• Reducing redundant queries by caching prior research results

Pros

• Lightweight TypeScript library with minimal dependencies
• Designed specifically for agent-to-agent memory sharing
• Open source with a permissive license

Cons

• Limited community adoption with only 8 GitHub stars
• No documented support for non-TypeScript environments
• Unclear integration complexity with existing agent frameworks

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.