byted-mediakit-shared
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the shell to execute the
mediakit-clitool for media processing, configuration management, and environment diagnostics. - [EXTERNAL_DOWNLOADS]: The instructions direct the user to install the
@volcengine/mediakit-clipackage from the npm registry to enable the skill's capabilities. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external media URLs and cloud task responses, representing a potential surface for indirect injection.
- Ingestion points: Media files are ingested from remote URLs (
http://,https://) or local paths, and task status is retrieved from cloud API responses (SKILL.md, reference/query_task.md). - Boundary markers: The tool utilizes structured JSON schemas for inputs and outputs to help maintain data integrity.
- Capability inventory: The skill possesses shell execution capabilities for the
mediakit-clitool and local filesystem access. - Sanitization: The documentation does not explicitly detail sanitization procedures for content fetched from remote URLs or API responses.
Audit Metadata