sumsub-theme-msdk

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill reads and processes data from untrusted mobile project source files and configuration (Swift, Kotlin, XML, JSON, Plist) to extract design tokens. This behavior creates a surface for indirect prompt injection if an attacker embeds malicious instructions within comments or metadata of the scanned project. • Ingestion points: File scanning and token extraction logic in 'ios/stages/1-scan.md' and 'android/stages/1-scan.md'. • Boundary markers: The skill does not implement delimiters or explicit 'ignore' instructions when reading host application code. • Capability inventory: The skill uses 'Write', 'Edit', and 'Bash' capabilities to create new files and modify project integration code. • Sanitization: No sanitization is performed on the extracted tokens before they are interpolated into the generated theme code.
  • [DYNAMIC_EXECUTION]: The skill generates Swift or Kotlin source code from templates and offers to run build commands (such as Gradle) to verify compilation. • Evidence: Logic in 'android/stages/3-generate.md' and 'ios/stages/3-generate.md' encourages the agent to invoke the project's build system to confirm the generated theme functions correctly.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 09:03 PM
Security Audit — agent-trust-hub — sumsub-theme-msdk