pov-vlog-presence-by-1798qo
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references external media URLs from
goodcase.aiandx.comwithinreferences/cases.md. These links are used by the agent to inspect finished media for visual verification when drafting new video prompts. This behavior is documented and aligns with the skill's primary purpose of cinematic style synthesis. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data from the
references/cases.mdfile and user-provided inputs to generate video prompts. - Ingestion points: The agent reads the contents of
references/cases.mdand user-supplied character/action requirements during every invocation. - Boundary markers: The skill mandates a 'Preserve / Replace / Avoid' reference contract to explicitly define which traits are inherited from the reference data versus user input.
- Capability inventory: The skill is limited to text generation for video prompts and does not possess capabilities for subprocess execution, file writing, or network operations beyond viewing the linked media.
- Sanitization: A verification checklist is implemented to ensure the output maps back to specific structural traits and avoids unrelated themes unless requested.
Audit Metadata