Search

More Filters

Show More
Add Listing
Go Back

evolving agents

0
New
Multi-Agent Platforms & Frameworks
Report

Report Abuse

Basic Information

Evolving Agents Toolkit (EAT) is a Python toolkit for building advanced, adaptive multi-agent systems that orchestrate discovery, creation, execution, and evolution of agents and tools to achieve high level goals. It centers on a SystemAgent orchestrator that plans and manages workflows and uses a unified MongoDB backend for persistent storage, vector search, agent registry, logs, caching, and intent plans. The repository contains core infrastructure such as SmartLibrary, SmartAgentBus, SmartContext, LLMCache, workflow engines, and example scripts including a comprehensive invoice processing demo. The project has been officially discontinued and archived in July 2025, with the core ideas simplified and reimplemented in a successor project called LLMunix and a new research organization called Evolving Agents Labs. The README is retained as archived documentation and for reference by developers and researchers.

Links

Stars
442
App URL
https://github.com/matiasmolinas/evolving-agents
Github Repository
https://github.com/matiasmolinas/evolving-agents/blob/main/README.md

App Details

Features
EAT provides a SystemAgent central orchestrator for goal driven planning and execution, a SmartLibrary that stores components with dual embeddings for content and applicability to enable task aware semantic search, and a SmartAgentBus that manages agent discovery, registration, and routing. The toolkit unifies persistence in MongoDB including vector search indexes, an LLMCache for completions and embeddings, an internal workflow engine with GenerateWorkflowTool and ProcessWorkflowTool, and an intent review system that supports human in the loop approval of plans. Smart Memory components were integrated, including a MemoryManagerAgent and tools for experience storage, semantic search, and summarization. Examples and demos are updated to use the MongoDB backend and show component evolution and orchestrated workflows.
Use Cases
EAT helps developers create scalable, governed multi-agent ecosystems by providing reusable components for discovery, versioning, semantic search, communication, and evolution. It enables AI centered workflows where a central agent issues plans, locates or creates tools, and executes tasks while persisting context and logs in MongoDB for reproducibility and audit. The dual embedding strategy improves retrieval of relevant components, the LLM cache reduces redundant calls, and the intent review features permit human oversight. Smart Memory integration enables recording and retrieval of past experiences to inform future decisions. The repo includes installation guidance, MongoDB setup notes, and runnable examples to demonstrate orchestration, though active development has moved to a simplified successor and this project is archived.

Categories

Content Creation (5042)
Monetization & Business (2246)
Development Tools (1407)
Marketing & Social Media (1403)
Education & Career (1380)
Multi-Agent Platforms & Frameworks (1009)
Health & Life (957)
AI Resources (920)
Agents & Automation (778)
Data & Analytics (764)
Productivity Tools (679)
Writing & Editing (666)

Similar Listings

listing image

virtual lab

0 (0 review)
listing image

mnemo

0 (0 review)
listing image

EdgeChains

0 (0 review)
listing image

RAG Agents Accelerator

0 (0 review)
listing image

LLM Zero to Hundred

0 (0 review)

Featured Listings

Tags

Please fill the required fields*