od-video-frame-logo-outro
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill utilizes the
$ARGUMENTSplaceholder to insert user-provided descriptions directly into a processing brief for theprepare_open_design_brieftool. - Ingestion points: User input is accepted via
$ARGUMENTSin theSKILL.mdfile. - Boundary markers: The user-controlled text is interpolated into a string without delimiters or specific instructions for the agent to treat the input as untrusted data.
- Capability inventory: The agent is instructed to 'Author the files yourself with your own file-editing tools', which involves file system write operations based on the instructions returned by the brief preparation tool.
- Sanitization: There is no evidence of sanitization, escaping, or validation performed on the user-provided input before it is passed to the next tool in the chain.
Audit Metadata