Engineering

XML Output Prompts

Legacy integrations and enterprise systems still speak XML. The output contract for well-formed, single-root XML responses — same engine, XML rules.

Open in JSON Output Prompt Builder

Overview

XML output is a compatibility play: SOAP-era integrations, enterprise middleware, feed formats. The contract engine is the same as JSON's — schema, example, validation, strictness — with XML's failure modes addressed by name: exactly one root element, one child element per field, entity escaping for &, <, and >. This resource loads an invoice-metadata contract in XML mode. One boundary worth knowing: this is XML as OUTPUT — using XML tags to structure your prompt's INPUT for Claude is a different technique, covered by the System Prompt Generator.

How to use this resource

Generate in XML mode

Note the single-root rule and the escaping rule — XML's two signature failures, addressed by name.
Check the empty-element convention

Unknown optional values become empty elements, keeping the element set stable for downstream XPath.
Keep element order

Strict mode pins schema order — XML consumers are far more order-sensitive than JSON ones.

Why This Works

Naming XML's failure modes (multiple roots, unescaped entities) prevents them specifically
A stable element set with empty-element nulls keeps XPath expressions valid across runs
The output/input boundary keeps this tool out of Claude-tag territory — different problem, different tool

Best for

Legacy system integrations where XML is non-negotiable
Middleware pipelines validating against XML structure
Enterprise environments mid-migration from XML to JSON

Not for

Structuring prompt INPUT with XML tags for Claude — that's input formatting, see the System Prompt Generator
Greenfield integrations — JSON parses cheaper; use the format select

Use cases

Feeding model output into SOAP-era and enterprise XML consumers
Generating feed-format entries that downstream XML parsers ingest
Porting an existing JSON contract to an XML-only integration

FAQ

How do I get well-formed single-root XML out of an AI model

The OUTPUT RULES pin exactly one <result> root element with one child element per field, and require escaping &, <, and > inside element values, naming XML's two signature failures. JSON Output Prompt Builder generates this contract in XML mode; you run the prompt in your own assistant, then validate the returned XML against your parser before ingesting it.

How are unknown optional values represented in the XML output

As empty elements, not omitted tags. The OUTPUT RULES say to use empty elements for unknown values, and VALIDATION RULES keep optional fields present but empty, so the element set stays stable across runs and downstream XPath keeps matching. In the invoice schema that means line_item_count appears as an empty element rather than disappearing when absent.

Is this for wrapping my prompt input in XML tags for Claude

No, that is a different technique. This is XML as OUTPUT, an invoice-metadata contract returning a <result> document; using XML tags to structure your prompt's INPUT is input formatting. The notFor points that work to the System Prompt Generator, keeping this tool out of Claude-tag territory and focused on the returned document instead.

Customize This Resource

Opens this setup in JSON Output Prompt Builder. Generate to get the full output contract — then adjust the schema, format, and strictness.

Open in JSON Output Prompt Builder

Prompt Template

Copy it as-is, or use Open in JSON Output Prompt Builder to load it pre-filled and customize it with your own context.

TASK
Read the invoice in the input and return its metadata as a structured record for our accounting system.

OUTPUT FORMAT
Return your entire response as a single well-formed XML document with <result> as the root element, matching the schema below.

SCHEMA
Each field, its type, and its meaning:
- invoice_number (string, required): Invoice reference number
- issue_date (string, required): Issue date, YYYY-MM-DD
- vendor (string, required): Issuing company name
- total (number, required): Grand total as a decimal number
- currency (string, required): ISO currency code
- line_item_count (number, optional): Number of line items

EXAMPLE OF A VALID RESPONSE
<result>
  <invoice_number>REC-1042</invoice_number>
  <issue_date>2026-06-10</issue_date>
  <vendor>Acme Corp</vendor>
  <total>49.99</total>
  <currency>USD</currency>
  <line_item_count>3</line_item_count>
</result>

VALIDATION RULES
- Every required field must be present.
- Optional fields with no value: include the element, leave it empty.
- Field types must match the schema: numbers without quotes, booleans as true/false.

OUTPUT RULES
- Return only the XML document — no text before or after it.
- Do not wrap the XML in markdown code fences.
- Use exactly one <result> root element; one child element per field.
- Escape &, <, and > inside element values; use empty elements for unknown values.

More resources from JSON Output Prompt Builder

Resource

Force JSON Output from AI

Stop getting 'Sure, here is the JSON…' — the output-contract pattern that forces models to return only parseable JSON: schema, example, and a strict rule block.

Prompt Engineering

Resource

JSON Mode Prompt — With and Without Native Support

Native JSON modes guarantee syntax, not your schema. The prompt contract that covers field names, types, and null discipline — whether or not the API has JSON mode.

Prompt Engineering

Resource

CSV Output Prompts

When the destination is a spreadsheet, CSV is the contract: one header row, one data row, quoting rules, and an honest answer about nested data.

Engineering

Resources that pair well

Resource

Extract Data From Text with AI

Free text in, named fields out. The extraction prompt pattern that turns any unstructured text into consistent, parseable records.

Prompt Engineering

Resource

Fix Invalid JSON from AI

The JSON won't parse and you can't see why. Deterministic cause-sniffing — trailing commas, single quotes, unclosed brackets — and the repair prompt that fixes it.

Engineering

Resource

Fix a Vague Prompt

'Make it good', 'be detailed', 'keep it interesting' — vague prompts get vague output. The fix is mechanical: replace every fuzzy word with a checkable instruction.

Prompt Engineering

Related tools

Tool

JSON Output Prompt Builder

Build prompts that return structured data — JSON first, with YAML, XML, and CSV modes — parseable every time.

Structured Output

Tip: Save time by exploring related resources and tools that integrate with this resource.