ai-translated-localization
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill involves an AI agent processing and translating strings from
Localizable.xcstringsfiles, which serves as a surface for indirect prompt injection. - Ingestion points: Untrusted data is ingested from external string catalog files during the translation fan-out pass.
- Boundary markers: No specific boundary markers or 'ignore' instructions are documented to separate the translation data from the agent's instructions.
- Capability inventory: The skill allows the agent to modify files in the repository (writing back to
.xcstrings) and use AI generation based on external inputs. - Sanitization: There are no mentioned mechanisms to sanitize or validate the content of the strings to be translated, which could allow adversarial text to influence the agent's output or actions.
Audit Metadata