# Vragen - **Website:** - **Primary alias:** `vragen.ai` ## Description Vragen.ai is an AI platform that helps organisations provide reliable, source-backed answers through conversations on their websites. It draws on an organisation’s own content, with administrators choosing which sources the AI can use and setting rules for its tone, response length and behaviour. Visitors can ask questions directly and receive answers with citations, while teams can manage and improve their AI agents through a central dashboard. The platform offers a website AI assistant and search functionality, alongside tools for reviewing conversations, evaluating answer quality and analysing common questions and themes. Its insights help organisations identify unanswered questions, weak responses and gaps in their content, so they can update information or refine agent instructions. Vragen.ai also supports integrations and provides features for testing and building agents. The company serves a range of sectors and organisations, including healthcare providers, government bodies, municipalities, businesses, agencies, publishers, museums, archives and recruitment teams. It highlights privacy, security, certifications and hosting as part of its offering. ## Industries - 🖥 Computers Electronics and Technology - 🖥 Artificial Intelligence and Machine Learning _(under Computers Electronics and Technology)_ ## Links - [linkedin](https://linkedin.com/company/vragen-ai) ## Logos & icons - logo _(primary)_ — PNG, SVG — [download](https://cdn.brandfetch.io/idPQVdsNwH/w/820/h/79/theme/dark/logo.png?c=1bxid64Mup7aczewSAYMX&t=1791372195242) - icon _(primary)_ — JPEG — [download](https://cdn.brandfetch.io/idPQVdsNwH/w/400/h/400/theme/dark/icon.jpeg?c=1bxid64Mup7aczewSAYMX&t=1791372195487) ## Colors | Name | Hex | Theme | | --- | --- | --- | | Torea Bay | #0e1ea3 | accent | | Tuna | #3f3f46 | dark | | White | #ffffff | light | ## Fonts - var(--font-kop) — asset _(custom)_ - var(--font-basis) — asset _(custom)_ --- _Source: _ _See [/llms.txt](/llms.txt) for a full list of agent-readable pages, and [/auth.md](/auth.md) for how agents authenticate._