Policy Catalog
WonderFence ships with a comprehensive catalog of pre-built detection models covering the most common AI safety and security risks. Each policy is backed by purpose-built models that are ready to enforce immediately — no training required to get started.
Policies are organized into the following groups:
| Group | Coverage |
|---|---|
| Security | Prompt injection, jailbreaking, system exploitation, data exfiltration attempts. |
| Privacy | PII detection (names, emails, phone numbers, addresses, financial data), sensitive data exposure. |
| Safety | Hate speech, violence, self-harm, sexual content, illegal activities, harassment. |
A rail on the left of the catalog scopes it. Under Category Presets, pick Security, Safety or Privacy to see only that category's policies; under Industry Presets, pick a preset to see the policies curated for that industry; under Frameworks, pick OWASP or MITRE to see every policy mapped to at least one risk in that framework. Each item shows how many policies it holds. The rail scopes the catalog to one item at a time, and clicking the selected item again shows the whole catalog.
To build your own preset, click Add Preset at the bottom of the rail, name it, tick the policies it runs and click Save. It appears under Custom Presets. Select a custom preset to Edit Preset (rename it or change its policies) or Delete Preset. A preset name must be unique among your project's WonderFence presets; a WonderBuild preset may share it.
To narrow further to individual risks, use the Risk Framework filter above the catalog: it lists each OWASP LLM Top 10 risk and MITRE ATLAS technique on its own, and shows the policies mapped to any of the risks you pick. It works together with the rail scope rather than replacing it.
The catalog can be viewed as a card grid or a list using the view toggle above it; your choice is remembered for next time. In both views, policies are grouped under a header per category, with a count of the policies in it; policies without a category are listed last. Each policy shows at a glance its name, description, source (System Policy or Custom Policy), its enforcement action (or a Rule-based pill in its place when the policy uses rule-based actions), the number of applications the policy is applied to (hover the count to see the application names), and its associated risk-framework indicators (OWASP LLM Top 10, MITRE ATLAS). A policy that is not applied to any application is grayed out in the catalog, so unused policies are easy to spot.