GHSA-c54g-xjwj-8g82
Hugo: XSS via text/html content files
Packages
| Ecosystem | Package | Affected | Fixed in |
|---|---|---|---|
| Go | github.com/gohugoio/hugo | < 0.162.0 | 0.162.0 |
Details
As written in the advisory (Markdown source, shown as plain text).
**Commit:** [e41a06447d](https://github.com/gohugoio/hugo/commit/e41a06447d) — _Disallow HTML content by default_
**Affected versions:** all Hugo versions prior to v0.162.0.
**Fixed in:** v0.162.0.
**Severity:** Low to Medium, depending on threat model. Not an issue if you fully trust every file under `/content` and every content adapter you load.
**Description.** Hugo accepts content files in several markup formats. Files mapped to the `text/html` media type (typically `.html` files under `/content`, or pages produced by a content adapter that sets `content.mediaType = "text/html"`) had their body emitted verbatim into the rendered page. A site that ingests HTML content from an untrusted source — for example, a CMS-backed editor, a content adapter pulling from an external API, or an automated import pipeline — could therefore be served stored cross-site scripting.
**Mitigation.** v0.162.0 introduces a `security.allowContent` whitelist with `text/html` denied by default. Sites that intentionally author HTML content can opt back in:
```toml
[security]
allowContent = ['.*']
```
This only affects pages whose source file (or content adapter output) declares an HTML media type; Markdown, AsciiDoc, Org, Pandoc and reStructuredText content is unaffected.
References
External links from the advisory. They lead to third-party sites.