claude-emporium  by Vvkmnn

Claude AI plugin marketplace and synergy framework

Created 3 weeks ago

New!

476 stars

Top 64.3% on SourcePulse

GitHubView on GitHub
Project Summary

This repository, Vvkmnn/claude-emporium, offers a plugin marketplace and architecture designed to extend Claude's capabilities. It targets Claude users and developers seeking to enhance AI workflows with modular, synergistic components, providing a zero-overhead, extensible ecosystem for features like context management, session memory, and tool discovery.

How It Works

The system utilizes a plugin-MCP (Microservice-like Component) architecture. Plugins act as thin wrappers, employing hooks to trigger MCP servers at critical junctures, thereby avoiding code duplication. MCP servers execute the core logic. A key design principle is "synergy," where installed plugins automatically detect and coordinate with each other at runtime, enhancing overall behavior without explicit configuration. This approach ensures zero overhead, with hooks running inline (~1ms) and computation/storage handled efficiently in-memory or project-locally.

Quick Start & Requirements

Installation involves adding the marketplace via /plugin marketplace add Vvkmnn/claude-emporium, followed by installing individual plugins using /plugin install claude-<plugin-name>@claude-emporium. Prerequisites include Node.js, typically available within the Claude Code environment. Standalone MCP servers can also be installed using claude mcp add <name> -- npx <package>.

Highlighted Details

  • Zero Overhead Architecture: Eliminates external API calls, network requests, and background processes. Hooks execute inline (~1ms), with computation and storage optimized for efficiency.
  • Runtime Synergy: Installed plugins automatically detect and coordinate with siblings, enhancing capabilities (e.g., Praetorian compacting research found by Historian).
  • Core Plugins: Offers six specialized plugins: Praetorian (context guard), Historian (session memory), Oracle (tool discovery), Gladiator (learning/adaptation), Vigil (file recovery), and Orator (prompt optimization).
  • Efficient Storage: Sensitive data like context snapshots, observations, and file checkpoints are stored in project-local .claude/ directories.
  • In-Memory Computation: Oracle, Historian, and Orator leverage in-memory processing with TTL caching for speed.

Maintenance & Community

The project welcomes contributions via Pull Requests for bug fixes, improvements, documentation, and new plugin ideas. Issue templates are provided for structured reporting and requests. The repository is explicitly marked as "[UNDER CONSTRUCTION]".

Licensing & Compatibility

The project is licensed under the MIT license, which permits broad use, including commercial applications and linking within closed-source projects.

Limitations & Caveats

The project is explicitly marked as "[UNDER CONSTRUCTION]", indicating it is not yet production-ready. It requires a Node.js environment, typically provided by Claude Code, and its functionality is confined to extending the Claude ecosystem.

Health Check
Last Commit

3 days ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
534 stars in the last 22 days

Explore Similar Projects

Starred by Eric Zhu Eric Zhu(Coauthor of AutoGen; Research Scientist at Microsoft Research), Elvis Saravia Elvis Saravia(Founder of DAIR.AI), and
2 more.

deepagents by langchain-ai

2.0%
10k
Framework for building advanced LLM agents
Created 7 months ago
Updated 1 day ago
Feedback? Help us improve.