web-prototype
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted source code and project configuration files (e.g., Xcode, Android, and web manifests) to generate design and logic documentation. This creates a surface for indirect prompt injection where malicious instructions embedded in the source project could influence the agent's behavior during implementation steps.
- Ingestion points: Native source project files (Swift, Kotlin), web manifest files (package.json, angular.json), and framework configurations.
- Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded commands when reading source files.
- Capability inventory: The agent is authorized to modify files in the prototype project and create/update local state files (design.md, logic.md, etc.) which persist across sessions.
- Sanitization: No sanitization or validation is performed on the data extracted from the source projects before it is used to populate documentation and plans.
Audit Metadata