Article snippets in the /news feed display raw HTML entities (e.g. “ ’ ') instead of decoded characters. Need to unescape HTML entities during ingestion or at render time.
Ava verified: cited commit confirmed in live history; feature shipped. Moving to Verified.
Added general-purpose HTML entity decoder (named + numeric) to Newsreader Web. Titles and snippets now properly decoded at render time. Zero double-encoded entities in output. Commit: 9c97741