Memory that learns and keeps itself current. A six-layer memory stack for Claude Code plus a nightly learning loop (capture, consolidation, scouts, conductor) that promotes your lessons into rules and surfaces new tools that fit your stack. Free, MIT.

personal skill pack

Pluggable session-log review and append-only event ledger

Multi-agent coding OS — BMad pipelines, Karpathy wiki, closed learning loops, dual-runtime coordination (Claude + Codex)

kimiflow — Claude Code + Codex plugin for codebase mapping, gated feature/fix flows, token-cheap learning memory, and Obsidian Vault MCP setup.

AI Powered assistant that generates football tweet drafts from news sources, delivers them via a Telegram bot, and learns from your engagement metrics.

completion-learn: post-task learning loop for finished work, ordered as self → collaboration → tool.

4-agent Python pipeline for CrowdWisdomTrading — competitor research, marketing report, Reddit pain-point finder, and closed learning loop using OpenRouter + Apify

Qomni Engine v8: Adaptive Cognitive OS — Crystal Distillation + CRYS-L v2 (open) + Learning Loop Planner · Percy Rojas Masgo, Qomni AI Lab

Self-learning loop skill for AI agents — cross-runtime (OpenClaw / Claude Code / Opencode / Codex)

A safe global learning loop for Codex on macOS.

🧬 Experience-driven continuous evolution skill for AI agents. Adapted from Hermes Agent (Nous Research) for OpenClaw.

Qomni Planner v0.59: Adaptive Cognitive Routing with Online Learning Loops for Edge AI Inference — Percy Rojas Masgo, CEO Condesi Perú · Qomni AI Lab

A cognitive layer for Claude Code — inspired by how brains actually work

Evidence-plane runtime for LawFirm OS: contract-pinned exception intake, fail-closed validation, append-only audit logs, and governed learning-loop candidates.

Zero-dependency multi-agent development workflow plugin for Claude Code