devexpress-xaf-filtering

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill describes a potential vulnerability surface where dynamic filtering criteria (via CriteriaOperator.Parse) could be built from untrusted data such as user input or external configuration.
  • Ingestion points: Specifically identifies that CriteriaOperator.Parse may receive strings from user input or configuration files (SKILL.md, references/criteria-syntax.md).
  • Boundary markers: The skill explicitly instructs the agent to use ? placeholders for parameterization and warns against string concatenation to prevent injection.
  • Capability inventory: The criteria strings are processed by IObjectSpace methods to execute queries against the application database.
  • Sanitization: Clear guidance is provided to use parameterized queries, which is the standard mitigation for this vector in XAF applications.
  • [SAFE]: The skill recommends using official DevExpress MCP tools for documentation retrieval. These tools facilitate access to verified API references and do not involve unauthorized code execution or network exfiltration.
  • [SAFE]: All NuGet packages (DevExpress.Data, DevExpress.ExpressApp, etc.) and domains (docs.devexpress.com) identified in the skill are official vendor-owned assets, representing normal and expected functionality for this framework.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 01:48 PM
Security Audit — agent-trust-hub — devexpress-xaf-filtering