Agentic Blueprints 131
Filters
Self-RAG
Retrieval with self-reflection at every step: grade the documents, check the answer for hallucination, and loop until it's grounded.
LLMCompiler
A planner streams a DAG of tasks, workers execute them in parallel as dependencies clear, and a joiner decides to finish or replan.
Code Assistant with Self-Correction
Generates code grounded in docs, executes it, and loops on the errors — reflecting on failures before retrying until tests pass.
Swarm Airline Service
OpenAI's airline demo: a triage agent hands customers to flight-change, cancellation, or lost-baggage agents, each with policy function tools.
SQL Agent
An agent that explores a database's schema, writes a query, double-checks it with a dedicated checker step, and recovers from execution errors.
Airline Customer Support Bot
A primary assistant hands off to specialized sub-assistants for flights, hotels, cars, and excursions, with user confirmation before writes.
Swarm Customer Service Streaming
A streaming customer-service pipeline from the Swarm repo, packaged with docker-compose and tests — the deployment-shaped example.
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.
Chatbot Simulation Evaluation
A simulated-user agent plays the customer against your chatbot so you can evaluate the whole conversation before real users do.
Corrective RAG
Grades retrieved documents for relevance and, when they fail the grade, falls back to web search before answering.
Swarm Support Bot
A two-agent support bot: a user-interface agent talks to the customer while a help-center agent answers from docs in a vector store.
WebVoyager
A vision-enabled browser agent that looks at annotated screenshots and decides where to click, scroll, and type until the task is done.
Self-Discover
The agent first composes its own reasoning structure from atomic modules, then solves the task by following the structure it built.
Agents SDK Research Bot
OpenAI's Agents SDK research flow: a planner derives searches, they run in parallel, and a writer synthesizes the report.
Agent Patterns Gallery
Sixteen single-file patterns from OpenAI: deterministic pipelines, agents-as-tools, parallelization, routing, LLM-as-judge, guardrails, human-in-the-loop.
Financial Research Agent
Six agents produce a verified financial research note: planner, searcher, fundamentals and risk analysts as tools, writer, then a verifier.
Agents SDK Customer Service
An airline customer-service team on the OpenAI Agents SDK: triage hands off to FAQ and seat-booking agents with function tools.
Stock Analysis Crew
Financial analyst, research analyst, and investment advisor agents dig through SEC filings and news to produce a full stock analysis.
Meeting Prep Crew
Researches the people and companies you're about to meet and compiles a briefing so you walk in knowing who's across the table.
Email Auto Responder Flow
A CrewAI Flow that checks an inbox on a loop and drafts responses for new email, saving them as drafts rather than sending.
Meeting Assistant Flow
Turns meeting notes into follow-through: a crew analyzes the meeting, then the flow files tasks and posts updates to your tools.
Self-Evaluation Loop Flow
One crew writes in Shakespeare's voice, a second crew reviews the output, and the flow loops the feedback until the writing passes.