homepage-video-assets
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill contains multiple shell command templates for
ffmpeg,avconvert, andremotionto handle video rendering and transcoding on the local system. - [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that interpolates external identifiers into shell commands, representing a surface for indirect prompt injection.
- Ingestion points: Variable placeholders like
<composition-id>,<asset>,<chrome-name>, and<safari-name>inSKILL.mdact as entry points for external data. - Boundary markers: There are no explicit instructions or markers provided to the agent to escape or sanitize these inputs before interpolation.
- Capability inventory: The skill utilizes local shell execution for media processing tools which have the capability to interact with the filesystem and execute sub-processes.
- Sanitization: The skill lacks any input validation, filtering, or sanitization logic for the data intended for the placeholders.
Audit Metadata