omni-recall

SKILL.md

Omni-Recall is a high-performance memory management skill designed for AI agents. It enables persistent, cross-session awareness by transforming conversation history and technical insights into high-dimensional vector embeddings, stored in a decentralized Supabase (PostgreSQL + pgvector) knowledge cluster.

Encodes current session state, user preferences, and operational steps into 1536-dimensional vectors using OpenAI's text-embedding-3-small via APIYI. Includes automatic duplicate detection (skips if cosine similarity > 0.9). Supports optional category and importance fields.

Pulls historical neural records from the last N days to re-establish the agent's mental model and context. Supports optional multiple keyword filtering (AND logic) and category filtering.

Omni-Recall : connaissances neuronales et moteur de contexte à long terme. Gère la mémoire de l'agent inter-session via Supabase (pgvector) et APIYI. Invoquer pour stocker/récupérer un contexte persistant. Source : ralph-wren/omni-recall.

Voir l'original