{"schemaVersion":"1.0","source":"The Workflow Finder","sourceUrl":"https://www.theworkflowfinder.com","license":"Free to use with attribution and a link to the cited page","generated":"2026-08-28","workflow":{"title":"Orchestrate a Multi-Agent Research Pipeline","slug":"orchestrate-a-multi-agent-research-pipeline","url":"https://www.theworkflowfinder.com/workflows/orchestrate-a-multi-agent-research-pipeline","description":"Split a research task across three specialized agents, planner, searcher, and synthesizer, so each one does less and does it more reliably than a single agent trying to do everything at once.","difficulty":"Advanced","timeEstimate":"3-4 hours, one-time setup","schemaDuration":"PT4H","expectedResult":"A working three-agent pipeline that takes a research question, plans sub-questions, retrieves grounded sources, and hands back a synthesized brief with citations, instead of one agent guessing across the whole task at once.","tags":["Agents","Research","Automation"]},"steps":[{"position":1,"title":"Split the Job Into Three Roles","instructions":"Define a Planner (breaks the research question into 3-5 sub-questions), a Searcher (retrieves grounded sources for each sub-question), and a Synthesizer (writes the final brief from what the Searcher found). Write down exactly what each role hands to the next: this handoff contract is what keeps the pipeline from collapsing into one confused agent.","input":"Workflow inputs: the topic, goal, or raw materials the user is starting from.","output":"Handoff artifact consumed by Step 2 (\"Build the Crew\").","ioProvenance":{"input":"structural","output":"structural"},"toolSlots":[]},{"position":2,"title":"Build the Crew","instructions":"Use CrewAI to define the three agents with their roles, goals, and the order they hand off work in. CrewAI's role-based design maps directly onto the Planner/Searcher/Synthesizer split, so you're configuring the pattern rather than fighting the framework to express it.","input":"Output of Step 1 (\"Split the Job Into Three Roles\").","output":"Handoff artifact consumed by Step 3 (\"Give the Searcher Real Retrieval, Not Just a Model's Memory\").","ioProvenance":{"input":"structural","output":"structural"},"toolSlots":[{"slug":"crewai","name":"CrewAI","url":"https://www.theworkflowfinder.com/tools/crewai","category":"Developer Tools","pricing":"free","capabilities":["Multi-step agent orchestration","Autonomous task execution"]}]},{"position":3,"title":"Give the Searcher Real Retrieval, Not Just a Model's Memory","instructions":"Wire the Searcher agent to Perplexity's API for live, cited web search instead of letting it answer from the model's training data. This is the single most common failure point in agent research pipelines: an ungrounded Searcher just hallucinates sources with confidence.","input":"Output of Step 2 (\"Build the Crew\").","output":"Handoff artifact consumed by Step 4 (\"Add a Verification Pass Before the Final Brief\").","ioProvenance":{"input":"structural","output":"structural"},"toolSlots":[{"slug":"perplexity","name":"Perplexity","url":"https://www.theworkflowfinder.com/tools/perplexity","category":"Research","pricing":"freemium","capabilities":["Web research","Source synthesis"],"hasApi":true}]},{"position":4,"title":"Add a Verification Pass Before the Final Brief","instructions":"Route the Synthesizer's draft through a separate Claude call whose only job is to check that every claim traces back to a source the Searcher actually retrieved, flagging anything that doesn't.","input":"Output of Step 3 (\"Give the Searcher Real Retrieval, Not Just a Model's Memory\").","output":"Handoff artifact consumed by Step 5 (\"Know Where This Still Needs a Human\").","ioProvenance":{"input":"structural","output":"structural"},"toolSlots":[{"slug":"claude","name":"Claude","url":"https://www.theworkflowfinder.com/tools/claude","category":"Writing","pricing":"freemium","capabilities":["Long-form writing","Code generation","Web research"],"hasApi":true}]},{"position":5,"title":"Know Where This Still Needs a Human","instructions":"This pipeline is reliable at gathering and organizing sources, it is not reliable at judging whether the Planner scoped the original question well. Skim the sub-questions before the Searcher runs, especially on the first few runs for a new topic: a poorly scoped plan produces a confidently wrong brief just as fast as a well-scoped one produces a good one.","input":"Output of Step 4 (\"Add a Verification Pass Before the Final Brief\").","output":"Final workflow output: A working three-agent pipeline that takes a research question, plans sub-questions, retrieves grounded sources, and hands back a synthesized brief with citations, instead of one agent guessing across the whole task at once.","ioProvenance":{"input":"structural","output":"structural"},"toolSlots":[]}]}