---
name: "query-reformulation-expander"
description: "Propose retrieval query rewrites and evaluate whether they improve relevant coverage. Use when the user needs: Get labeled rewrites, a retrieval comparison and unsupported-claim flags."
metadata: {"version": "2026-10-17.1", "category": "search-retrieval", "publisher": "AIDiveForge"}
---

# Query Reformulation Expander

Get labeled rewrites, a retrieval comparison and unsupported-claim flags.

## Requirements

Provide the original query, corpus/index interface and relevance criteria.

## Workflow

1. Extract the user’s actual information need and required constraints before adding terminology.
2. Produce a few meaning-preserving rewrites plus a clearly labeled hypothetical answer only if the retrieval system supports it.
3. Run the original and rewrites under the same retrieval settings and preserve result IDs.
4. Deduplicate results and assess relevance against the original need, not against the hypothetical answer.
5. Report observed coverage and false positives; claim improved recall only with suitable labeled relevance evidence.

## Worked example

Read `examples/input.txt` and `examples/expected.txt` before adapting the workflow.

## Acceptance checks

Preserve the side-effect constraint and label the hypothetical passage as generated, not sourced evidence.

## Limits

Rewrites can reduce precision. Without relevance judgments, a larger result set is not proof of improved recall.

Worked example included. Model-specific performance and vendor compatibility have not been benchmarked.