Open-source causal graph memory for AI agents. 89.9% on LoCoMo. MCP server with ACT-R scoring, spreading activation, and active forgetting.

EMNLP'22, CEM improves MHCH performance by correcting prediction bias and training an auxiliary cost simulator based on user state and labor cost causal graph, without requiring complex model crafting.

This is an official repository for "Exploring Multi-Modal Data with Tool-Augmented LLM Agents for Precise Causal Discovery".

Local-first long-term memory for AI coding tools — causal graphs, hybrid BM25+vector search, 2.46x context augmentation

An application of causal inference methods to optimizing the location of delivery drivers.

Append-only event DAG prototypes with RECENT_A trace-closure frontier + conflict-pair bound signals.

A graph-based failure modeling and deterministic detection system for LLM agent runtimes.

🧠 Implement a bi-factual contrastive explanation system for AI decisions, enhancing understanding through formal definitions and optimized algorithms.

Canonical Ontology: A closed causal language of reality. Engineering protocol mapping the 25-node graph from 1. Causality to 25. Universe. [Genso & Akane]

Implémentation d’un système d’IA Explicable (XAI) basé sur les explications contrastives bi-factuelles, avec optimisations algorithmiques et interface graphique CausaLytics.

S.A.G.A. — Story Analysis, Generation, and Archives. Canon-aware narrative intelligence for analysis, retrieval, timelines, and story generation.

Graph-first desktop workbench for debugging multi-agent runs — understand any execution in 30 seconds

Industry - Casualty Challenge

An MCP server that gives AI agents a permanent, append-only causal history