Agentic Blueprints 14
Filtersactive
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.
LATS (Language Agent Tree Search)
Monte-Carlo tree search over agent actions: generate candidates, reflect and score them, expand the best branches until one solves the task.
Reflection (LangGraph tutorial)
The essay-writing reflection loop: a generator drafts, a reflector critiques as a teacher grading, and the loop repeats a fixed number of rounds.
Reflexion
An answerer that critiques itself with citations: draft, self-review for missing and superfluous content, search to fill gaps, revise.
Plan-and-Execute (LangGraph tutorial)
The original plan-and-execute notebook: a planner writes the step list, an executor works through it, and a replanner revises as results come in.
WebVoyager
A vision-enabled browser agent that looks at annotated screenshots and decides where to click, scroll, and type until the task is done.
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.