Agentic Blueprints 29
Filtersactive
Self-Evolving Agent
An EvoAgentX system that generates its own multi-agent workflow for a goal, executes it, verifies the result, and repairs itself.
Multi-Agent Debate
Several solver agents answer the same problem, read each other's answers, and revise over rounds until the majority converges.
Mixture of Agents
Layers of worker agents refine the previous layer's answers, with an orchestrator combining them into a stronger final response.
Group Chat (AutoGen pattern)
A group-chat manager coordinates role agents — writer, editor, and friends — publishing turns to a shared conversation.
Sequential Workflow (AutoGen pattern)
Four agents in a strict pipeline, each transforming the previous agent's output — the assembly-line shape, minimal and clean.
Semantic Router
A semantic router classifies each incoming message and dispatches it to the right agent over a distributed pub/sub runtime.
Negotiation Battle Simulator
Two agents negotiate against each other in a simulated deal — agent-vs-agent with a live UI, built on Google ADK and AG-UI.
Magentic-One
Microsoft's generalist five-agent team: an orchestrator directing a web surfer, file surfer, coder, and terminal to solve open-ended tasks.
Selector Group Chat
An LLM selector reads the conversation and picks which agent speaks next — planner, searcher, or analyst — instead of a fixed turn order.
Distributed Group Chat
A writer and an editor collaborate in a group chat where every agent runs in a separate distributed runtime, talking over gRPC.
Handoffs (AutoGen pattern)
Swarm-style handoffs in AutoGen: triage, sales, and refund agents transfer the conversation to whoever should own it next.
Concurrent Agents (AutoGen pattern)
Fan a task out to multiple agents at once and gather their results — AutoGen's minimal parallel pattern.
Reflection (AutoGen pattern)
A coder agent and a reviewer agent in a critique loop: code, review, revise, until the reviewer approves.
Chatbot Simulation Evaluation
A simulated-user agent plays the customer against your chatbot so you can evaluate the whole conversation before real users do.
Hierarchical Agent Teams
A supervisor of supervisors: a top-level orchestrator directs a research team and a document-writing team, each with its own internal supervisor.
Adaptive RAG
Routes each question to the right retrieval strategy, grades what comes back, and falls back to web search when the index can't answer.
Corrective RAG
Grades retrieved documents for relevance and, when they fail the grade, falls back to web search before answering.
Self-Discover
The agent first composes its own reasoning structure from atomic modules, then solves the task by following the structure it built.
Self-RAG
Retrieval with self-reflection at every step: grade the documents, check the answer for hallucination, and loop until it's grounded.
Multi-Agent Collaboration
A researcher agent and a chart-generator agent divide a task and route work between themselves until the answer is a finished chart.
LLMCompiler
A planner streams a DAG of tasks, workers execute them in parallel as dependencies clear, and a joiner decides to finish or replan.
ReWOO
Reasoning without observation: a planner writes the full tool plan up front with variable substitution, workers fill it in, a solver answers.