chatcut-media-guide
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided media files (video, audio, images), which presents a surface for indirect prompt injection via embedded metadata or file content.\n
- Ingestion points: User-selected local or attached media files as described in SKILL.md.\n
- Boundary markers: The instructions explicitly require the agent to explain the cloud upload boundary and confirm user agreement before importing.\n
- Capability inventory: The agent uses the 'ChatCut import tool', accesses the project editor, and manages the project timeline (SKILL.md).\n
- Sanitization: There are no explicit instructions for filtering or sanitizing file content or metadata.\n- [NO_CODE]: The skill package contains no executable scripts, binaries, or configuration files, and consists purely of markdown instructions and metadata.
Audit Metadata