feat(agent): scrape linked pages to ground answers in real content

Adds `_fetch_page_content()`, which downloads a URL with a
browser-like User-Agent and uses BeautifulSoup with the lxml
parser to strip scripts, navigation, and other non-content
markup, returning up to 3000 characters of readable text.

The response validator now also compares generated replies
against this scraped text, flagging named entities and claims
that do not appear in the retrieved page content.

Dependencies: adds beautifulsoup4==4.12.3 and pins lxml==5.3.0
for parser compatibility.
This commit is contained in:
2026-10-26 16:48:29 -03:00
parent 9eed02645b
commit 723ca1506b
231 changed files with 20508 additions and 5330 deletions
@@ -34,11 +34,7 @@ typedef enum {
XML_SAVE_XHTML = 1<<4, /* force XHTML1 specific rules */
XML_SAVE_AS_XML = 1<<5, /* force XML serialization on HTML doc */
XML_SAVE_AS_HTML = 1<<6, /* force HTML serialization on XML doc */
XML_SAVE_WSNONSIG = 1<<7, /* format with non-significant whitespace */
/* Available since 2.14.0 */
XML_SAVE_EMPTY = 1<<8, /* force empty tags, overriding global */
XML_SAVE_NO_INDENT = 1<<9, /* disable indenting */
XML_SAVE_INDENT = 1<<10 /* force indenting, overriding global */
XML_SAVE_WSNONSIG = 1<<7 /* format with non-significant whitespace */
} xmlSaveOption;
@@ -77,27 +73,17 @@ XMLPUBFUN int
xmlSaveFlush (xmlSaveCtxtPtr ctxt);
XMLPUBFUN int
xmlSaveClose (xmlSaveCtxtPtr ctxt);
XMLPUBFUN xmlParserErrors
xmlSaveFinish (xmlSaveCtxtPtr ctxt);
XMLPUBFUN int
xmlSaveSetIndentString (xmlSaveCtxtPtr ctxt,
const char *indent);
XML_DEPRECATED
XMLPUBFUN int
xmlSaveSetEscape (xmlSaveCtxtPtr ctxt,
xmlCharEncodingOutputFunc escape);
XML_DEPRECATED
XMLPUBFUN int
xmlSaveSetAttrEscape (xmlSaveCtxtPtr ctxt,
xmlCharEncodingOutputFunc escape);
XML_DEPRECATED
XMLPUBFUN int
xmlThrDefIndentTreeOutput(int v);
XML_DEPRECATED
XMLPUBFUN const char *
xmlThrDefTreeIndentString(const char * v);
XML_DEPRECATED
XMLPUBFUN int
xmlThrDefSaveNoEmptyTags(int v);