openshield

openshield is a hosted API for moderation and fraud prevention, run by Sway SRL in Belgium. Send an image or a text from your server and get accept, review or reject, with the categories, scores and thresholds behind it: nudity, gore, weapons, hate symbols, hate speech, personal data, spam, scams and your own image categories. The same call checks what your users post and what your AI agents read and write. Send a sign-up, login or payment event and get allow, step up, hold or block, with the rules that fired; pack rules start in shadow and act on their own after a 30-day replay. Content is checked in memory and never stored. Built on openmod and openfraud, open-source Rust engines (Apache-2.0). Pay per check from a prepaid balance, with up to 2,000 free checks a month and no card.

openshield provides APIs for content moderation and fraud prevention. Its moderation API checks images and text and returns an accept, review, or reject decision with categories and scores. Its fraud prevention API scores sign-ups, logins, and payments and returns an allow, step up, hold, or block decision with the reasons that fired. The service helps teams moderate user and agent content before publication and assess fraud risk before sensitive actions. Moderated content is checked in memory and not stored. A user creates an account, gets a default workflow, and creates an API key in the dashboard. The application sends image bytes or text to the moderation API, or sends an event such as an account creation to the fraud API. The response includes a decision and its reasons; moderation responses can include category scores and references. Workflows let users set category thresholds and versions. A coding agent can use the integration prompt, documentation, and OpenAPI document to add checks to code. The page presents openshield for SaaS products that moderate posts, comments, avatars, and uploads, and for teams moderating content sent to or generated by AI agents. It also serves organizations scoring sign-ups, logins, and payments for fraud. Notable details include a shared account for moderation and fraud prevention, EU moderation servers and databases, and fraud instances in Germany. The hosted service is built on open-source engines.