cef-cubby-state

Store and manage persistent cubby state for CEF agents via the Cubby API.

1|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/CEF-AI/skills --skill cef-cubby-state
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cef-cubby-state
Source: https://github.com/CEF-AI/skills/tree/main/skills/storage
Command: npx skills add https://github.com/CEF-AI/skills --skill cef-cubby-state

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Cubbies provide a centralized, persistent storage layer for CEF agents, enabling durable state, per-entity isolation, and shared state across agents via migrations and a SQLite-backed API.

Core Features & Use Cases

  • Cubby API with SQLite query/exec, migrations, and multiple instances
  • State machine pattern, deduplication, buffered flush, and sqlite-vec vector storage
  • Safe reads and per-entity isolation across agent workloads

Quick Start

Configure a cubby alias in cef.config.yaml, redeploy your engagement, and start persisting per-entity state using ctx.cubbies and the Cubby API.

Frequently Asked Questions about cef-cubby-state

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I persist per-entity state for CEF agents using SQLite?

Persist per-entity state for CEF agents by configuring a cubby alias in cef.config.yaml, then utilizing ctx.cubbies and the Cubby API to execute SQLite queries, run migrations, and maintain isolated durable state.

What is the best way to manage cross-agent state sharing without data conflicts?

Manage cross-agent state sharing safely by using per-entity cubby instances with explicit instanceId usage, which enforces per-entity isolation and prevents data conflicts across concurrent agent workloads.

Does SQLite work with vector storage for agent state management?

SQLite supports vector storage for agent state management through the sqlite-vec extension, enabling you to store and manage vector data within the persistent cubby state layer alongside traditional relational data.

How do I run database migrations to define schemas for agent state?

Run database migrations to define schemas for agent state by using the migrations-defined schemas approach within the Cubby API, which requires inline handler code and explicit instanceId usage to ensure production reliability.

Can I use state machine patterns and deduplication with buffered flush in SQLite?

You can implement state machine patterns, deduplication, and buffered flush within SQLite using the Cubby API, which provides a centralized, persistent storage layer designed for safe cross-agent sharing and per-entity isolation.

Why do I need explicit instanceId usage when querying persistent cubby state?

Explicit instanceId usage is required when querying persistent cubby state to guarantee per-entity isolation and safe reads, preventing data leakage across concurrent CEF agent workloads in production deployments.