cueframe-product-video
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches and runs the vendor's official CLI tool using 'npx -y cueframe@0.12' from the npm registry. This is standard behavior for the skill's purpose and uses a version-pinned package from a well-known service.
- [COMMAND_EXECUTION]: Instructs the agent to execute shell commands for video composition, validation, and rendering (e.g., 'cueframe composition put', 'cueframe render'). These commands are part of the intended product functionality and are limited to the vendor's tooling.
- [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface for processing untrusted external data within video compositions.
- Ingestion points: Video caption text and overlay labels provided in the composition JSON ('text', 'subtitle', and 'words' fields).
- Boundary markers: Relies on JSON structure; no specific instructions provided to ignore potential injection content in strings.
- Capability inventory: The 'render' command which processes the JSON into an MP4 file.
- Sanitization: No explicit sanitization or validation of the input text is described in the workflow.
Audit Metadata