od-video-frame-liquid-bg-hero
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to author files using its file-editing tools based on instructions returned from the
prepare_open_design_brieftool. This creates a potential vector for indirect prompt injection if the external service's output contains malicious instructions. - Ingestion points: Output from the
prepare_open_design_briefMCP tool. - Boundary markers: Absent; the skill does not use delimiters or provide warnings to the agent to treat the external tool output as untrusted data.
- Capability inventory: The agent is authorized to use file-editing tools and register artifacts based on the tool's instructions.
- Sanitization: No validation or sanitization of the tool's returned instructions is performed before execution.
- [COMMAND_EXECUTION]: The skill requires the agent to perform file system operations (authoring files) that are dynamically determined by instructions from a remote source.
Audit Metadata