maple-chatcut-story-first-editing
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious instructions, obfuscation, or unauthorized execution patterns were found. The skill follows best practices by instructing the agent to never invent tools or install unrelated dependencies.
- [SAFE]: Project security is prioritized; the instructions mandate that project and timeline IDs must be kept in private notes rather than shared, and that connection states must be validated before any mutation occurs.
- [INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface involving transcripts and content from external brand websites (e.g., for color verification). Ingestion points include transcripts, timeline clips, and external brand sites. Boundary markers include required project state read-back and manual frame inspection. The capability inventory involves mutation of project timelines, audio, captions, and effects via MCP. Sanitization is achieved through verification checklists and human-in-the-loop review handoffs. Given the extensive verification steps and human-in-the-loop requirements, this surface is considered low risk and assessed as safe.
Audit Metadata