openmotion-production
Pass
Audited by Gen Agent Trust Hub on Nov 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources, which creates a standard vulnerability surface for indirect prompt injection.
- Ingestion points: Processes user-supplied briefs, approved copy, and local/URL-based media assets as described in
references/intake.md. It also reads and measures external reference videos as specified inreferences/reference-video.md. - Boundary markers: No explicit boundary markers or instructions to ignore embedded commands within processed data were identified in the provided instructions.
- Capability inventory: The skill possesses significant capabilities across multiple files, including project file writing (
openmotion_edit_projectinreferences/native.md), shell command execution via Node.js scripts (references/reference-video.md,references/fonts.md), and rendering/export operations (openmotion_export_requestinreferences/verification.md). - Sanitization: The instructions do not describe explicit sanitization or filtering of external content before it is processed or interpolated into the production workflow.
- [EXTERNAL_DOWNLOADS]: The skill references and fetches assets from external sources.
- Evidence: The
fonts/catalog.jsonfile contains numerous URLs for downloading font binaries fromraw.githubusercontent.com/google/fontsandgithub.com/lxgw/LxgwWenKai. These resources are fetched via theprepare-font.tsscript mentioned inreferences/fonts.mdto set up the local production environment. These downloads target trusted organizations and well-known open-source projects. - [COMMAND_EXECUTION]: The skill utilizes shell commands to perform technical tasks.
- Evidence: Instructions in
references/fonts.mdandreferences/reference-video.mddirect the agent to execute Node.js scripts within the runtime environment (e.g.,node scripts/prepare-font.ts,node scripts/font-specimen.ts, andnode <hyperframesCli> measure). These commands are used for font preparation, specimen generation, and video measurement. - [DYNAMIC_EXECUTION]: The skill facilitates the generation and execution of content at runtime.
- Evidence:
references/native.mddescribes a 'Native Hyperframes authoring' process where the agent writes HTML, CSS, and Javascript to a local project directory (openmotion_edit_project) which is then executed or rendered by the OpenMotion engine. This is the core functionality of the skill for producing video content.
Audit Metadata