matrixscan-batch-ios
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read existing project source code and write integration code back to the file system, which constitutes a standard attack surface for developer-oriented AI agents.
- Ingestion points: The skill instructions (in SKILL.md and references/migration.md) require the agent to read user-provided Swift files and dependency manager lockfiles like Package.resolved and Podfile.lock.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the ingested project data as untrusted or to ignore any instructions embedded within it.
- Capability inventory: The skill is granted high-autonomy capabilities to write code directly to the project files ("write the integration code directly into that file") rather than just displaying it in the chat.
- Sanitization: There are no requirements for the agent to sanitize or validate the content of the project files before processing or modifying them.
- [EXTERNAL_DOWNLOADS]: The skill references and provides instructions for fetching the Scandit Data Capture SDK from the official vendor repository on GitHub (https://github.com/Scandit/datacapture-spm) and utilizes technical documentation hosted on official Scandit domains (docs.scandit.com).
Audit Metadata