video
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read marketing context from external files such as
.agents/product-marketing-context.mdor.claude/product-marketing-context.md. - Ingestion points: The agent is prompted to read these local files at the start of the workflow in
SKILL.md. - Boundary markers: No specific delimiters or instructions to ignore embedded commands within these files are provided.
- Capability inventory: The skill possesses command execution capabilities (e.g.,
npm install,npx) and can interact with external APIs. - Sanitization: There are no explicit instructions to sanitize or validate the content retrieved from these context files before using them to drive the video production logic.
- [COMMAND_EXECUTION]: The skill includes instructions to execute shell commands for setting up development environments.
- Evidence: Mentions of
npm install hyperframesandnpx create-video@latestinSKILL.mdfor framework initialization. - [EXTERNAL_DOWNLOADS]: The skill references and downloads tools from established external services and package registries.
- Evidence: References to official documentation and tools from well-known services like HeyGen, Runway, and Remotion in
SKILL.mdandreferences/ai-video-prompting.md.
Audit Metadata