What caching mechanisms are used for performance?

The manager uses a multi-level caching strategy (L1/L2/L3) combined with predictive prefetching based on access patterns and LRU eviction for maximum memory efficiency.

Does this skill support data persistence?

Yes, it includes write-ahead logging, automated checkpointing, and point-in-time recovery to ensure that the swarm's memory is durable and recoverable.

What is the primary purpose of the Swarm Memory Manager?

It serves as a distributed consciousness keeper that manages collective memory across multiple agents, ensuring data consistency and efficient retrieval through advanced synchronization protocols.

Can I monitor the health of the memory swarm?

Yes, the skill writes detailed performance metrics every 60 seconds, including operations per second, cache hit rates, sync latency, and memory usage.

How does it handle data conflicts between different agents?

It implements Conflict-free Replicated Data Types (CRDTs), vector clocks for causality tracking, and versioned last-write-wins logic to resolve conflicts automatically.

Swarm Memory Manager

Name: Swarm Memory Manager
Author: nbossn

bynbossn

0•

Database Management

Coordinates distributed memory and state synchronization across multi-agent systems to ensure data consistency and high-performance retrieval.

The Swarm Memory Manager acts as the central consciousness keeper for agent hive minds, implementing advanced synchronization protocols like CRDTs and vector clocks to manage shared memory. It provides a robust framework for multi-level caching, atomic writes, and conflict resolution, ensuring that all agents in a swarm operate on a single source of truth. By automating memory persistence and performance tracking, it enables complex, long-running agent workflows to maintain context without data loss or state corruption across distributed nodes.

Key Features

010 GitHub stars

02Automatic checkpointing and point-in-time recovery procedures

03Multi-level (L1/L2/L3) caching with predictive prefetching

04Real-time performance monitoring and memory health metrics

05Atomic write coordination with automated conflict resolution

06Distributed state synchronization using CRDTs and vector clocks

Use Cases

01Implementing reliable state persistence for complex, long-running multi-step AI workflows

02Maintaining shared context and global variables across a swarm of autonomous coding agents

03Optimizing data retrieval latency in high-concurrency agent environments through advanced caching

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nbossn/claude-code-base agent-swarm-memory-manager

For use in Claude.ai and ChatGPT

Key Features

010 GitHub stars

02Automatic checkpointing and point-in-time recovery procedures

03Multi-level (L1/L2/L3) caching with predictive prefetching

04Real-time performance monitoring and memory health metrics

05Atomic write coordination with automated conflict resolution

06Distributed state synchronization using CRDTs and vector clocks

Use Cases

01Implementing reliable state persistence for complex, long-running multi-step AI workflows

02Maintaining shared context and global variables across a swarm of autonomous coding agents

03Optimizing data retrieval latency in high-concurrency agent environments through advanced caching

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nbossn/claude-code-base agent-swarm-memory-manager

For use in Claude.ai and ChatGPT