{"id":7480,"date":"2026-08-19T19:10:08","date_gmt":"2026-08-19T19:10:08","guid":{"rendered":"https:\/\/www.wiven.ai\/wiven-llm\/docs\/administration\/controle-de-contenu\/"},"modified":"2026-08-19T19:10:08","modified_gmt":"2026-08-19T19:10:08","slug":"controle-de-contenu","status":"publish","type":"wvn_doc","link":"https:\/\/www.wiven.ai\/en\/wiven-llm\/docs\/administration\/controle-de-contenu\/","title":{"rendered":"Content control"},"content":{"rendered":"<h1>Content control<\/h1>\n<p>Content control applies <strong>rules<\/strong> to user messages and<br \/>\nto the assistant&#039;s responses. It serves to enforce internal policy:<br \/>\nprevent the entry of data that has no place in a conversation.,<br \/>\nto conceal sensitive information, to exclude topics outside the scope.<\/p>\n<blockquote>\n<p>Function disabled by default. <code>Settings \u2192 Content Control<\/code>.<\/p>\n<\/blockquote>\n<hr \/>\n<h2>How is a rule defined?<\/h2>\n<p>Each rule includes:<\/p>\n<ul>\n<li>A <strong>name<\/strong>, which appears in the monitoring of offences; ;<\/li>\n<li>A <strong>pattern<\/strong> : either a comma-separated list of terms, or a<br \/>\n  regular expression for structured cases; ;<\/li>\n<li>a <strong>breakage sensitivity<\/strong>, can be activated; ;<\/li>\n<li>a <strong>scope<\/strong> : incoming messages, outgoing replies, or both; ;<\/li>\n<li>a <strong>action<\/strong>.<\/li>\n<\/ul>\n<h3>The shares<\/h3>\n<p><strong>To block.<\/strong> The message is rejected and an explanation message is displayed. Nothing.<br \/>\nis not transmitted to the model.<\/p>\n<p><strong>Hide.<\/strong> The corresponding passages are replaced by a mark of<br \/>\nMasking, and processing continues. Useful for letting a question pass.<br \/>\nwhile removing sensitive data.<\/p>\n<p><strong>Report.<\/strong> Treatment continues normally, but the occurrence is<br \/>\nsaved. The right way to test a rule before actually applying it.<\/p>\n<hr \/>\n<h2>Write useful rules<\/h2>\n<h3>By list of terms<\/h3>\n<p>Each term is searched as a whole word \u2014 a short term will not trigger<br \/>\nso not on a syllable within another word. A term containing<br \/>\nspaces is sought as a complete expression.<\/p>\n<p>Suitable for vocabulary such as: confidential project names, internal designations,<br \/>\ntopics outside scope.<\/p>\n<h3>By regular expression<\/h3>\n<p>Suitable for the following formats: structured numbers, internal identifiers, references<br \/>\nstandardized email addresses from a given domain.<\/p>\n<blockquote>\n<p>An invalid expression is not applied. Test your patterns, and begin<br \/>\nthrough action <strong>Report<\/strong> to observe what they actually capture before<br \/>\nto block.<\/p>\n<\/blockquote>\n<hr \/>\n<h2>The displayed message<\/h2>\n<p>A default message explains that content has been blocked by the policy of<br \/>\nthe organization. It is customizable at the instance level, and each rule<br \/>\ncan also carry its own message.<\/p>\n<p><strong>Write it for a user, not for an administrator.<\/strong> \u00ab&quot;This message<br \/>\ncontains data that should not be entered here. Contact department X for<br \/>\n&quot;The procedure to follow&quot; is useful; &quot;Violation of rule 7&quot; is not.<\/p>\n<hr \/>\n<h2>Monitoring of offences<\/h2>\n<p>Each trigger is logged with the relevant rule, the user, the<br \/>\ndate and a <strong>short excerpt of context<\/strong> \u2014 never the entire message.<\/p>\n<p>This monitoring allows us to:<\/p>\n<ul>\n<li>identify a rule that is too wide and hinders the work; ;<\/li>\n<li>spot a ruler that is too narrow and lets through what it should catch; ;<\/li>\n<li>document the incidents.<\/li>\n<\/ul>\n<hr \/>\n<h2>What content control is not<\/h2>\n<p><strong>This is not protection against exfiltration.<\/strong> A determined user<br \/>\nbypasses a pattern filter. Content control is used to <strong>prevent the<br \/>\nerrors<\/strong>, not the intentions.<\/p>\n<p><strong>This is not a semantic moderation.<\/strong> It works on patterns, not on<br \/>\nThe meaning. A roundabout way of phrasing it works.<\/p>\n<p><strong>This is not a substitute for access rights.<\/strong> What should not be seen<br \/>\nmust not be located in the workspace.<br \/>\n\u2192 <a href=\"\/en\/wiven-llm\/docs\/administration\/utilisateurs\/\">Users and roles<\/a><\/p>\n<hr \/>\n<h2>Recommended implementation<\/h2>\n<ol>\n<li><strong>Start by observing.<\/strong> Create your rules in action <em>Report<\/em> and leave<br \/>\n   filming for a few days.<\/li>\n<li><strong>Analyze the tracking.<\/strong> How many triggers? Legitimate or false?<br \/>\n   positive?<\/li>\n<li><strong>Adjust the patterns<\/strong>, then proceed to <em>Hide<\/em> Or <em>To block<\/em>.<\/li>\n<li><strong>Communicate.<\/strong> An unknown blocking rule is being experienced by users.<br \/>\n   like a breakdown.<\/li>\n<li><strong>Re-examine periodically.<\/strong> An organization&#039;s vocabulary evolves.<\/li>\n<\/ol>","protected":false},"parent":7451,"menu_order":505,"template":"","meta":[],"class_list":["post-7480","wvn_doc","type-wvn_doc","status-publish","hentry"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.6 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Contr\u00f4le de contenu \u2014 Documentation WivenLLM<\/title>\n<meta name=\"description\" content=\"Bloquer ou masquer certains contenus dans les messages et les r\u00e9ponses, et suivre les infractions.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.wiven.ai\/en\/wiven-llm\/docs\/administration\/controle-de-contenu\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Contr\u00f4le de contenu \u2014 Documentation WivenLLM\" \/>\n<meta property=\"og:description\" content=\"Bloquer ou masquer certains contenus dans les messages et les r\u00e9ponses, et suivre les infractions.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.wiven.ai\/en\/wiven-llm\/docs\/administration\/controle-de-contenu\/\" \/>\n<meta property=\"og:site_name\" content=\"Wiven AI\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"3 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.wiven.ai\\\/wiven-llm\\\/docs\\\/administration\\\/controle-de-contenu\\\/\",\"url\":\"https:\\\/\\\/www.wiven.ai\\\/wiven-llm\\\/docs\\\/administration\\\/controle-de-contenu\\\/\",\"name\":\"Contr\u00f4le de contenu \u2014 Documentation WivenLLM\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.wiven.ai\\\/#website\"},\"datePublished\":\"2026-08-19T19:10:08+00:00\",\"description\":\"Bloquer ou masquer certains contenus dans les messages et les r\u00e9ponses, et suivre les infractions.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.wiven.ai\\\/wiven-llm\\\/docs\\\/administration\\\/controle-de-contenu\\\/#breadcrumb\"},\"inLanguage\":\"en\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.wiven.ai\\\/wiven-llm\\\/docs\\\/administration\\\/controle-de-contenu\\\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.wiven.ai\\\/wiven-llm\\\/docs\\\/administration\\\/controle-de-contenu\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/www.wiven.ai\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Documentation\",\"item\":\"https:\\\/\\\/www.wiven.ai\\\/wiven-llm\\\/docs\\\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Administration\",\"item\":\"https:\\\/\\\/www.wiven.ai\\\/wiven-llm\\\/docs\\\/administration\\\/\"},{\"@type\":\"ListItem\",\"position\":4,\"name\":\"Contr\u00f4le de contenu\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.wiven.ai\\\/#website\",\"url\":\"https:\\\/\\\/www.wiven.ai\\\/\",\"name\":\"Wiven AI\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.wiven.ai\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Content Control \u2014 WivenLLM Documentation","description":"Block or hide certain content in messages and replies, and track violations.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.wiven.ai\/en\/wiven-llm\/docs\/administration\/controle-de-contenu\/","og_locale":"en_US","og_type":"article","og_title":"Contr\u00f4le de contenu \u2014 Documentation WivenLLM","og_description":"Bloquer ou masquer certains contenus dans les messages et les r\u00e9ponses, et suivre les infractions.","og_url":"https:\/\/www.wiven.ai\/en\/wiven-llm\/docs\/administration\/controle-de-contenu\/","og_site_name":"Wiven AI","twitter_card":"summary_large_image","twitter_misc":{"Est. reading time":"3 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/www.wiven.ai\/wiven-llm\/docs\/administration\/controle-de-contenu\/","url":"https:\/\/www.wiven.ai\/wiven-llm\/docs\/administration\/controle-de-contenu\/","name":"Content Control \u2014 WivenLLM Documentation","isPartOf":{"@id":"https:\/\/www.wiven.ai\/#website"},"datePublished":"2026-08-19T19:10:08+00:00","description":"Block or hide certain content in messages and replies, and track violations.","breadcrumb":{"@id":"https:\/\/www.wiven.ai\/wiven-llm\/docs\/administration\/controle-de-contenu\/#breadcrumb"},"inLanguage":"en","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.wiven.ai\/wiven-llm\/docs\/administration\/controle-de-contenu\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/www.wiven.ai\/wiven-llm\/docs\/administration\/controle-de-contenu\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.wiven.ai\/"},{"@type":"ListItem","position":2,"name":"Documentation","item":"https:\/\/www.wiven.ai\/wiven-llm\/docs\/"},{"@type":"ListItem","position":3,"name":"Administration","item":"https:\/\/www.wiven.ai\/wiven-llm\/docs\/administration\/"},{"@type":"ListItem","position":4,"name":"Contr\u00f4le de contenu"}]},{"@type":"WebSite","@id":"https:\/\/www.wiven.ai\/#website","url":"https:\/\/www.wiven.ai\/","name":"Wiven AI","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.wiven.ai\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en"}]}},"_links":{"self":[{"href":"https:\/\/www.wiven.ai\/en\/wp-json\/wp\/v2\/wvn_doc\/7480","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wiven.ai\/en\/wp-json\/wp\/v2\/wvn_doc"}],"about":[{"href":"https:\/\/www.wiven.ai\/en\/wp-json\/wp\/v2\/types\/wvn_doc"}],"up":[{"embeddable":true,"href":"https:\/\/www.wiven.ai\/en\/wp-json\/wp\/v2\/wvn_doc\/7451"}],"wp:attachment":[{"href":"https:\/\/www.wiven.ai\/en\/wp-json\/wp\/v2\/media?parent=7480"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}