---
id: ch5-agent-builder
type: spine
title: "Boxes & Arrows: build an agent instead of asking one question"
readingTime: 2
standalone: true
core: false
teaser: "A RAG chat reads 10–20 chunks, not the whole library. The Agent Builder in JackDaw turns a broad task into steps with checks."
parent: null
diagram: fig-agent-builder
recallQ: "Why does a single broad question to a RAG chat give weak answers, and what does the Boxes & Arrows Agent Builder change?"
recallA: "RAG matches the question against indexed chunks and returns only a small top-ranked set, often 10–20, so a broad question pulls up general statements. The Agent Builder lets users draw multi-step agents whose steps have their own instructions, knowledge and outputs, with branching, loops and quality checks."
highlights:
  - "RAG is a relevance match on a few chunks, not an analysis of the whole corpus"
  - "Each box is a step with its own instructions, background data and defined output"
  - "Branching, looping, merging and quality checks; made for non-technical users"
status: accepted
concept: polirural-advisor
lang: en
lens: generic
depth: standard
formalism: none
visuality: balanced
lengthBand: standard
genre: explainer
state: edited
carriers: prose|diagram
---

In the Beta, many users treated the JackDaw chat like a search box: one question, one final answer (D4.2 §3.6). Used that way, a conversational agent gives far less than it can.

## Why one broad question falls short

A RAG engine compares the question, or an expanded version of it, with indexed document chunks and returns a small top-ranked set, **often 10 to 20 chunks**. That is a relevance match against a small subset of the documents, **not an analysis of the whole corpus**. A specific, well-scoped question can be answered well. A broad question without clear limits tends to pull up general statements with little substance.

So the quality of the answer depends heavily on how the question is scoped. Some tasks, such as research or assessment work with known steps, prior knowledge and quality criteria, need a structured conversation rather than a single prompt.

## What the builder does

The **Boxes & Arrows Agent Builder**, a module of [JackDaw](#c/jackdaw), lets users draw an agent as a graph:
- each **box** is a step with its own instructions, background data or knowledge resources, and a defined output;
- **arrows** can branch, loop and merge;
- **quality checks** can sit between steps.

This lets complex research and assessment tasks be structured and partly automated. It is designed so that non-technical users can configure their own agents, which opens agentic AI to a wider group of users.

## Status

The builder is a new module in the Beta. D4.2 does not report pilot results for it. More support for model builders is one of D4.2's priorities before the final release: demos, workflow templates and configuration examples. The candidate demos include rural attractiveness analysis, stakeholder mapping and RAP support (D4.2 §11.1).

**Sources:**
- D4.2 PoliRuralPlus Platform Beta, §3.6 Boxes & Arrows Agent Builder; §11.1 Remaining development priorities. [PDF](https://www.poliruralplus.eu/docs/421/D4.2_PoliRuralPlus_Platform_Beta.pdf)