gamaze-labs/hicortex

Overview

Self-learning memory for AI agents. Captures agent experiences automatically, distills them into lessons overnight, and shares those lessons across a fleet of agents. Supports Hermes, OpenClaw, Claude Code, and Pi. Written in TypeScript.

Best for

Best for
Teams running fleets of AI agents that need shared, persistent memory.

Use cases

• Persist agent learnings across sessions
• Share distilled lessons across multiple agents
• Automate memory consolidation for agent fleets

How to use

Install

npx @gamaze/hicortex init

Tested with

Claude Desktop, Claude Code, Cursor, ChatGPT

Notes

Self-learning memory for AI agents. Captures agent experiences automatically, distills them into lessons overnight, and shares those lessons across a fleet of agents. Supports Hermes, OpenClaw, Claude Code, and Pi. Written in TypeScript.

2 stars on GitHub. Last updated 2026-09-15.

Use cases

• Persist agent learnings across sessions
• Share distilled lessons across multiple agents
• Automate memory consolidation for agent fleets

Pros

• Automatic capture and distillation of agent experience
• Cross-agent sharing of learned lessons
• Works with several popular agent frameworks

Cons

• Very early stage with only 2 GitHub stars
• Overnight distillation implies delayed availability of lessons
• Limited to supported agents (Hermes, OpenClaw, Claude Code, Pi)

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