swift-quiz
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill instructs the agent to use a Bash subagent to execute the
swiftcommand on code snippets provided by the user. This constitutes runtime execution of dynamically generated content incorporating external input. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface where untrusted data (user-written Swift code) is processed by a tool with shell execution capabilities.
- Ingestion points: User-provided Swift code in "Type D: Write Code" questions (SKILL.md).
- Boundary markers: None explicitly defined to delimit user code from the execution environment.
- Capability inventory: Accesses the Task tool with
subagent_type: "Bash"to run compiler commands. - Sanitization: No specific instructions are provided to sanitize or validate the user-supplied code before it is passed to the Swift compiler.
Audit Metadata