shotstack
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill includes instructions to install and update the
@shotstack/clipackage from the npm registry and usesnpx skills updatefor skill maintenance. - [COMMAND_EXECUTION]: The skill operates primarily by executing shell commands using the
shotstackCLI to orchestrate video rendering, asset generation, and studio previews. - [DATA_EXFILTRATION]: The
ingest uploadcommand facilitates the transmission of local media files to Shotstack's vendor-managed AWS S3 buckets to host them for rendering. This behavior is consistent with the primary purpose of the tool. - [INDIRECT_PROMPT_INJECTION]: The skill processes Edit JSON files and AI generation prompts from external sources, which constitutes an attack surface for instructions embedded in untrusted data.
- Ingestion points: Commands such as
shotstack render,shotstack validate, andshotstack ingestread local JSON files and external URLs. - Boundary markers: The skill recommends using a published OpenAPI schema and the
shotstack validatetool to check JSON structure and property names before processing. - Capability inventory: The skill utilizes subprocess execution of the
shotstackCLI and can launch a headless browser via theinspectcommand. - Sanitization: The skill relies on Shotstack's API-side schema validation and does not implement additional client-side sanitization of prompt data.
Audit Metadata