The OWASP AISVS Default Findings Engine is a requirement-aware template layer for OWASP Artificial Intelligence Security Verification Standard (AISVS). It helps a team define consistent starting language for recurring findings while the normal project remains the source of truth for scope, evidence, ownership, remediation progress, validation, and reports.
Library and standards scope
OWASP AISVS provides requirements for evaluating security controls around AI applications and their surrounding systems. Relevant evidence may involve prompts, retrieved content, model and data handling, agent tools, authorization, output processing, supply chains, logging, and abuse response. A model response alone rarely describes the whole application risk.
Global, local, and project layers
- Global defaults are standards-based templates published by voiqq platform owners for a system Library.
- Local defaults belong to the signed-in workspace administrator and can override editable wording without changing the global source.
- A project finding receives copied template values and then becomes an independent record.
- Later global changes do not silently rewrite a local override or a finding already created in a project.
- Status is Open and validation is Pending when the reusable default is applied, unless an authorized project workflow later changes them.
Fields and canonical mapping
Each template is owned by a canonical AI security requirement in the OWASP AISVS Library. The reusable record contains a stable identifier, summary, description, remediation guidance, severity key, mapped values, source Library, and display order. The visible wording can be edited locally, while the background requirement identity continues to support filters, reports, imports, and New Finding suggestions.
Create or customize a default
- Define the AI system type, model boundary, tools, data flows, and authorized test scope.
- Open the AISVS local engine and choose the most precise requirement.
- Describe the reusable application-control failure rather than one surprising response.
- State the security consequence, preconditions, and expected protective behavior.
- Add remediation across application, tool, data, policy, or monitoring controls as appropriate.
- Keep real prompts, outputs, accounts, models, repetitions, and evidence in the project finding.
Writing rules for reusable findings
Write the summary as a concise statement of the recurring failure. Use the description to explain the expected behavior, likely impact, or control concern in neutral language. Use remediation to describe the desired outcome rather than a patch tied to one framework or customer. Store actual results, reproduction steps, affected assets, evidence, people, dates, measurements, samples, and environment details in the project finding.
Suitable template subjects
- Untrusted instructions override higher-priority application controls through direct or indirect prompt injection.
- An agent invokes a privileged tool without checking the authenticated caller and requested resource.
- Sensitive information enters prompts, retrieval context, model output, or retained provider data unnecessarily.
- Model output reaches an interpreter, browser, workflow, or downstream system without safe handling.
- AI security events lack the prompt, policy, tool, model, and response context needed for investigation.
Use a default in a project
Open New Finding inside a project configured with OWASP AISVS. Choose Template mode or select a prepared template after choosing the applicable AI security requirement. voiqq prefills the reusable values. Review every field, add the real evidence and context, and save the finding. Comments, attachments, assignments, history, validation, sharing, exports, and reports then use the same normal project workflow.
Accuracy and safety checks
- Do not place sensitive prompts, outputs, API keys, or personal data in a default.
- Distinguish model behavior from application authorization and integration failures.
- Require repeatability and context before setting final severity.
- Map to the most specific supported AISVS requirement.
- Retest the complete application path, not only the model.
voiqq uses the stable OWASP AISVS 1.0 catalogue: 191 requirements in 12 chapters with verification levels 1, 2, and 3. AISVS requirements are assessment requirements, not prewritten findings; several findings can be connected to one requirement when the evidence warrants it.