walkthrough-video
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to install dependencies, execute browser automation commands via
playwright-cli, and run video rendering processes usingnpx remotion. These commands are standard for the tool's intended purpose. - [EXTERNAL_DOWNLOADS]: The skill installs several packages from the official NPM registry, including
remotion,react, andtypescript. These are well-known and trusted libraries in the JavaScript ecosystem. - [REMOTE_CODE_EXECUTION]: The skill generates TypeScript React code at runtime and executes it locally to produce video files. This dynamic execution is a core feature but represents a known execution pattern.
- [PROMPT_INJECTION]: The skill is subject to indirect prompt injection (Category 8) because it ingests data from external sources and user-provided manifests.
- Ingestion points: Captures data from external URLs via Playwright and reads metadata (titles, descriptions) from the
screens.jsonfile. - Boundary markers: The instructions do not define delimiters or markers to isolate potentially malicious instructions within the captured content.
- Capability inventory: The skill has access to the
Bashtool for running commands and theWritetool for generating executable code files. - Sanitization: There is no explicit logic described for sanitizing input data before it is interpolated into shell commands or generated code.
Audit Metadata