intent-sync
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill maintains a significant attack surface by scanning the entire codebase to extract implementation details for synchronization.
- Ingestion points: The skill reads original Intent files and scans all implemented code, including public interfaces, data models, and configuration files (e.g.,
src/,config/). - Boundary markers: The instructions do not specify the use of boundary markers or delimiters when processing code content to distinguish between code data and potential embedded instructions.
- Capability inventory: The skill utilizes file-read capabilities across the codebase and file-write capabilities to update the project's Intent documentation.
- Sanitization: There is no evidence of sanitization or escaping of content extracted from the codebase before it is interpolated into the updated documentation, which could lead to persistent indirect injections in the design files.
Audit Metadata