Custom Agent Creation Through Inheritance And Composition

via “pre-built agent patterns with llm-powered reasoning and code execution”

Microsoft's multi-agent framework — event-driven, typed messages, group chat, AutoGen Studio.

Unique: Provides a unified Agent interface where AssistantAgent, CodeExecutorAgent, WebSurferAgent, and FileSurferAgent all implement the same protocol, enabling them to be composed into teams without adapter code. Each agent type encapsulates domain-specific logic (LLM calls, subprocess execution, web scraping) while exposing a consistent message-based interface, allowing developers to swap implementations or add custom agents.

vs others: More composable than LangGraph's node-based approach because agents are first-class runtime objects with consistent interfaces; more flexible than CrewAI's role-based agents because agents can be dynamically instantiated and reconfigured at runtime without role definitions.

via “multi-agent orchestration with hierarchical agent types”

Google's agent framework — tool use, multi-agent orchestration, Google service integrations.

Unique: Implements three distinct agent execution patterns (Loop, Sequential, Parallel) as first-class types with explicit state hierarchy and context propagation, rather than generic agent composition. Each pattern has dedicated configuration classes (LoopAgentConfig, SequentialAgentConfig, ParallelAgentConfig) that enforce pattern-specific semantics and prevent misuse.

vs others: More structured than LangGraph's flexible graph approach — enforces specific execution semantics upfront, reducing debugging complexity for common multi-agent patterns at the cost of less flexibility for custom topologies

via “agent team composition with role-based specialization”

Microsoft AutoGen multi-agent conversation samples.

Unique: Agents are composed as independent instances with configurable tools and prompts, enabling true specialization; BaseGroupChat routes messages based on agent capabilities rather than fixed turn order

vs others: More modular than monolithic multi-agent frameworks because each agent is independently configurable and can be tested/debugged in isolation before team composition

via “micro-agent-decomposition-and-composition”

What are the principles we can use to build LLM-powered software that is actually good enough to put in the hands of production customers?

Unique: Implements agent decomposition as a first-class architectural pattern, with explicit composition mechanisms and DAG-based orchestration, rather than building monolithic agents that handle multiple concerns

vs others: More maintainable and testable than monolithic agents because each micro-agent has a single responsibility and can be tested/iterated independently, improving overall system reliability by 30-50% through reduced cognitive load

via “composable multi-plugin agent orchestration with tool routing”

The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra

Unique: Uses a standardized plugin interface with T5 format streaming for structured tool call handling, allowing plugins to be composed dynamically without tight coupling. The architecture separates agent orchestration logic from tool implementation, enabling independent scaling and testing of each plugin.

vs others: More modular than monolithic agent frameworks (like LangChain agents) because plugins are independently deployable and can run in isolated environments, versus frameworks that require all tools to be registered in a single process.

via “agent registry and multi-agent orchestration”

The Frontend Stack for Agents & Generative UI. React + Angular. Makers of the AG-UI Protocol

Unique: Implements agent registry as a runtime service that manages agent lifecycle and routing. Enables multiple agents to coexist in the same runtime with isolated state and tool execution contexts, supporting agent composition and delegation patterns.

vs others: More structured than ad-hoc agent selection; AgentRegistry provides centralized agent management and isolation. Enables agent composition patterns (one agent delegating to another) without custom orchestration code.

via “modular-component-system-capability-extension”

[GenAI Application Development Framework] 🚀 Build GenAI application quick and easy 💬 Easy to interact with GenAI agent in code using structure data and chained-calls syntax 🧩 Use Event-Driven Flow *TriggerFlow* to manage complex GenAI working logic 🔀 Switch to any model without rewrite applicat

Unique: Implements a ComponentSystem where agent functionality is extended through pluggable components (EventListener, Tool, Role) registered with agents rather than subclassing, with components coordinating through a shared RuntimeContext, enabling true composition-based agent design.

vs others: More flexible than LangChain's tool binding (which is function-focused) and cleaner than LlamaIndex's agent subclassing approach, with explicit component types (EventListener, Tool, Role) making intent clearer and enabling better code organization.

via “declarative agent composition and template instantiation”

Hi HN,I’m Vincent from Aden. We spent 4 years building ERP automation for construction (PO/invoice reconciliation). We had real enterprise customers but hit a technical wall: Chatbots aren't for real work. Accountants don't want to chat; they want the ledger reconciled while they slee

Unique: Provides declarative agent templates with parameterized behavior, allowing runtime instantiation of agent variants without code changes

vs others: More flexible than hardcoded agent factories, but requires learning framework-specific template syntax unlike generic dependency injection containers

via “registry-driven agent composition with hierarchical delegation”

AI agent framework for plan-first development workflows with approval-based execution. Multi-language support (TypeScript, Python, Go, Rust) with automatic testing, code review, and validation built for OpenCode

Unique: Uses a declarative registry.json as the single source of truth for agent definitions, enabling agents to be discovered and composed dynamically at runtime rather than through hardcoded imports. The hierarchical delegation pattern (primary agents → subagents) is explicitly modeled in the registry with typed component categories (Agents, Subagents, Contexts, Commands), allowing the framework to enforce composition rules and validate agent relationships during installation.

vs others: More maintainable than agent frameworks that require code changes to add new agents, and more flexible than monolithic agent designs because agents can be versioned, swapped, and composed independently through registry metadata rather than tight coupling.

via “extensible agent architecture with custom agent creation”

TradingAgents: Multi-Agents LLM Financial Trading Framework

Unique: Provides extensible agent architecture where custom agents can be created by extending base classes and implementing agent-specific logic, then registered in LangGraph graph. Agents receive state as input and produce outputs added to shared state, enabling seamless integration without modifying core framework.

vs others: More extensible than fixed-agent systems because it allows adding custom agents without framework changes. More flexible than generic agent frameworks because it provides trading-specific base classes and patterns that reduce boilerplate for financial agents.

via “nested agent hierarchies and agent composition”

Multi-agent framework with diversity of agents

Unique: Implements agent composition through a delegation pattern where parent agents can spawn or coordinate sub-agents, with automatic message routing and result aggregation. Supports both sequential and parallel sub-agent execution with configurable synchronization and error handling.

vs others: More structured than flat multi-agent systems because it enables clear task hierarchies and specialization, and more flexible than rigid workflow engines because agent hierarchies can be defined dynamically based on task requirements

via “custom agent behavior through inheritance and overrides”

Framework for orchestrating role-playing agents

Unique: Enables low-level customization through class inheritance and method overrides, allowing developers to modify core agent behavior while maintaining crew integration

vs others: More flexible than configuration-based customization but requires more expertise than role-based agent definition

via “crew composition and agent team assembly”

JavaScript implementation of the Crew AI Framework

Unique: Provides a declarative crew composition API where agents are assembled with explicit role assignments and tool bindings, enabling teams to be defined as configuration rather than code and reused across projects

vs others: More structured than ad-hoc agent creation because it enforces team composition patterns, but less flexible than fully dynamic agent networks

via “agent configuration and orchestration with yaml/json policy files”

Local-first personal agentic OS and everything app for coding, knowledge work, web design, automations, and artifacts.

Unique: Provides declarative YAML/JSON-based agent configuration with built-in orchestration and agent composition support, allowing non-technical users to define and route between agents without code, with capability-based access control integrated into configuration schema

vs others: More accessible than code-based agent definition for non-technical users, though less flexible than programmatic APIs for complex conditional logic or dynamic behavior

via “agent composition and hierarchical task decomposition”

We’ve been working with automating coding agents in sandboxes as of late. It’s bewildering how poorly standardized and difficult to use each agent varies between each other.We open-sourced the Sandbox Agent SDK based on tools we built internally to solve 3 problems:1. Universal agent API: interact w

Unique: Provides first-class support for agent composition with automatic state passing, error handling, and result aggregation, enabling hierarchical agents without manual orchestration logic

vs others: More integrated than manual agent orchestration because it handles state passing, error handling, and result aggregation automatically, reducing boilerplate compared to building composition logic manually

via “agent mixin composition for shared behavior and methods”

A fast and minimal framework for building agentic systems

Unique: Leverages Python's multiple inheritance and mixin pattern to compose agent capabilities, allowing @action-decorated methods from mixins to be automatically discovered and exposed without requiring explicit registration or configuration

vs others: More Pythonic than composition-based approaches (like wrapping agents) because it uses native language features; simpler than plugin systems because mixins are resolved at class definition time rather than runtime

via “extensible plugin architecture for custom agents”

[COLM 2024] OpenAgents: An Open Platform for Language Agents in the Wild

Unique: Uses a 'one agent, one folder' directory structure with automatic plugin discovery and shared adapters, enabling developers to add custom agents by implementing a standard interface without modifying core code

vs others: More modular than monolithic frameworks but requires more boilerplate than decorator-based plugins; enables code reuse through shared adapters but less flexible than fully composable agent patterns

via “agent composition and hierarchical task decomposition”

AI agent orchestration framework for TypeScript/Node.js - 29 adapters (LangChain, AutoGen, CrewAI, OpenAI Assistants, LlamaIndex, Semantic Kernel, Haystack, DSPy, Agno, MCP, OpenClaw, A2A, Codex, MiniMax, NemoClaw, APS, Copilot, LangGraph, Anthropic Compu

Unique: Provides framework-agnostic agent composition with automatic dependency resolution and parallel execution, allowing agents from different frameworks to be composed into hierarchies

vs others: Supports cross-framework agent composition (LangChain agents with CrewAI agents) unlike framework-specific composition; automatic dependency resolution reduces manual orchestration code

via “agent composition and nested agent orchestration”

Hi HN,Over Thanksgiving weekend I wanted to build an AI agent. As a design exercise, I wrote it as a set of React components. The component model made it easier to reason about the moving parts, composability was straightforward (e.g., reusing agents/tools), and hooks/state felt like a rea

Unique: Treats agents as React components that can be nested and composed like any other component, enabling agent hierarchies to be expressed as component trees with natural prop and context flow

vs others: More natural composition than external agent orchestration frameworks because agent composition is just React component composition, leveraging existing React patterns and tooling

via “extensible agent framework with custom agent creation”

Multi-agent general purpose platform

Unique: Provides a base agent class and shared adapter infrastructure that custom agents inherit, reducing boilerplate and ensuring consistency — developers implement only agent-specific logic while inheriting streaming, memory, and LLM integration automatically

vs others: More structured than building agents from scratch and more flexible than fixed agent types, though with less documentation than frameworks like LangChain that provide more detailed extension guides