filmcraft-editor
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local command (
node ui/dist/adapter.mjs --serve) to start the editor's backend service if it is not already active. - [INDIRECT_PROMPT_INJECTION]: The skill's workflow involving the ingestion of external assets and the ability to write to the workspace creates a surface for indirect prompt injection.
- Ingestion points: Materials and assets are loaded via the
open_filecommand into the workspace. - Boundary markers: There are no instructions or delimiters provided to help the agent distinguish between data and potential malicious instructions within processed files.
- Capability inventory: The skill allows for local command execution (via Node.js) and file writing (
save_file,save_project) across the workspace. - Sanitization: No sanitization or validation steps are defined for the external data imported by the agent.
Audit Metadata