AgentSkillsCN

Secure Xml Parsing

安全解析 XML

SKILL.md

SKILL: Secure XML Parsing for Untrusted Content

Confidence: low Source: earned Domain: .NET / C# / Security

When to Apply

When parsing XML files whose content originates from untrusted or semi-trusted sources (user uploads, CI artifacts, API responses, file downloads from external systems).

The Pattern

csharp
var settings = new XmlReaderSettings
{
    DtdProcessing = DtdProcessing.Prohibit,
    XmlResolver = null,
    MaxCharactersInDocument = 50_000_000 // ~50MB character limit
};

using var reader = XmlReader.Create(stream, settings);
var doc = XDocument.Load(reader);

Why This Matters

  1. •

    XXE (XML External Entity) attacks — Default XDocument.Load(stream) allows <!ENTITY> declarations that can read local files (file:///etc/passwd) or make HTTP requests to internal services (SSRF). DtdProcessing.Prohibit blocks all DTD processing.

  2. •

    Billion Laughs (XML bomb) — Nested entity expansion can turn a 1KB XML file into gigabytes of memory consumption. DtdProcessing.Prohibit prevents this. MaxCharactersInDocument provides a secondary defense.

  3. •

    External resolver SSRF — Even with DTDs disabled, XmlResolver could resolve external schema references. Setting it to null prevents all external resource resolution.

Checklist

  • • DtdProcessing = DtdProcessing.Prohibit — always
  • • XmlResolver = null — always
  • • MaxCharactersInDocument set to a reasonable limit — for untrusted input
  • • File size check BEFORE parsing (don't load a 500MB file into XDocument)
  • • Content truncation on extracted text fields (error messages, descriptions) to prevent downstream resource exhaustion (e.g., LLM context window overflow)

Common Mistakes

  • •Using XDocument.Load(path) or XDocument.Load(stream) directly — these use default settings which allow DTDs
  • •Setting DtdProcessing = DtdProcessing.Ignore instead of Prohibit — Ignore still processes some DTD constructs
  • •Forgetting XmlResolver = null — DTD prohibition alone doesn't prevent all external resolution paths
  • •Not checking file size before parsing — XDocument loads the entire DOM into memory

References