19 free toolsno accountnothing stored
Validate the XML, convert it, and find out exactly why it broke.
19 tools that run libxml2 on the server, not a regular expression in your browser. When one rejects a document, 322 reference pages explain what it meant.
Up to 2 MB per document. No signup, no upload, no retention.
<channel><item><title>First</title></channel>
Opening and ending tag mismatch: item line 2 and channel (line 4)
- Xerces (Java)
- The element type "item" must be terminated by the matching end-tag "</item>".
- expat (Python)
- mismatched tag: line 4, column 2
The toolbox
19 tools, and not one of them wants your email.
Paste a document, get an answer. Each tool parses in memory on the server and returns a result — no document, fragment, or filename is written to a database or a log.
Validate
Is this document acceptable — and if not, where exactly.
Convert
Move between XML and the shapes the rest of the stack speaks.
Inspect & clean
Read the structure, compare two versions, normalise the text.
The reference
A tool tells you it failed. This tells you why.
322 entries, every example run through the same parser the tools use before it was published — so the message on the page is the message you will get.
- 58
Glossary
The words the specs assume you know
- 24
Formats
Root element, namespaces, what breaks
- 49
Elements
Per-element rules, broken and fixed
- 33
Namespaces
URIs and their conventional prefixes
- 40
XPath 1.0
Functions, axes and operators that ran
- 39
XSD datatypes
What validates, and what only looks like it
- 29
Parse errors
Keyed by the message you were given
- 10
Schema errors
The ones only validation surfaces
- 31
Public endpoints
Real XML, re-fetched daily
- 9
Comparisons
XML or JSON, RSS or Atom, XSD or RNG
Practical guides
From the error message to the working document.
Focused walkthroughs for parser failures, feeds, sitemap incidents, namespace-aware XPath, XSD contracts, and SOAP faults.
- 6 min
XML
How to fix an invalid XML entity reference
Find the unescaped ampersand, distinguish named entities from character references, and repair the value without hiding malformed input.
- 7 min
Sitemaps
How to fix “Sitemap could not be read”
A deterministic checklist for separating fetch failures, HTTP problems, malformed XML, namespace mistakes, and invalid sitemap entries.
- 8 min
Languages
Validate XML in JavaScript and TypeScript
Build a layered JavaScript validation path that separates XML syntax, namespace and vocabulary diagnostics, XSD validation, and application rules.
- 12 min
Security
Prevent SAML XML signature wrapping
Validate SAML structure, signatures, destination, audience, time, request correlation, and replay without consuming an unsigned sibling assertion.
- 7 min
XSLT
Make XSLT match a default namespace
Fix XSLT templates that never fire when source elements use a default namespace, with patterns that work across XSLT versions.
- 9 min
Integrations
Validate XML in GitHub Actions with SARIF
Call the XMLDir validation API from a workflow, preserve filenames, fail the job on errors, and upload SARIF for pull-request annotations.
In practice
Built for the XML that has consequences.
Sitemap operations
Track sitemap freshness, broken URL patterns, missing lastmod values, oversized files, and indexability regressions.
Keep sitemap.xml healthy at scaleVendor integrations
Map the payloads partners actually send and catch contract drift before it breaks orders, claims, listings, or reports.
Catch contract drift before it breaks ordersXML to data workflows
Extract repeatable XPath maps and prepare XML for JSON, CSV, SQL, warehouse, or Postgres ingestion, with field mappings that survive upstream schema changes.
Turn XML into clean, repeatable pipelinesSample payload vault
Keep redacted examples, test fixtures, validation notes, and known quirks in a workspace teams can search, instead of in one person's inbox.
A searchable memory for tricky XML
The workspace
When one answer is not enough, keep the document.
The tools are stateless on purpose. A free workspace is the other half: it stores every version of a document, indexes its structure, diffs the versions against each other, and re-checks monitored URLs on a schedule.
No card. The guest workspace needs no account at all.
Indexed, not just uploaded
Every document becomes a namespace-aware map of paths, attributes, sample values, and schema references you can search.
Versions that answer questions
Keep each fetch of a document and compare them by structure, so "what changed" has an answer that is not a text diff.
Sources that report themselves
Point the monitor at a feed, sitemap, or partner URL and hear about malformed documents, missing fields, and shape changes.
What a stored record holds
- XML, XSD, DTD, RELAX NG, RSS, Atom, SOAP, sitemap.xml
- Namespace-aware indexing and XPath extraction
- Private workspaces, redaction-ready samples
- Source URL, fetch time, and verdict kept with every version