redbook-creator-publish
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the
playwrightpackage from PyPI and downloads thechromiumbrowser. These are established tools from well-known vendors (Microsoft and Google) used for automation. - [PROMPT_INJECTION]: The skill processes untrusted data from external web searches to inform its content generation, creating a vulnerability to indirect prompt injection.
- Ingestion points: Output from the
WebSearchtool used during the topic research phase. - Boundary markers: None are explicitly defined to isolate external search results from instructions.
- Capability inventory: Subprocess execution of local Python scripts for image generation and automation; write access to the local file system; network access via the automated browser.
- Sanitization: No specific filtering or escaping of the retrieved search content is mentioned before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill executes local system commands and Python scripts using
python3,pip3,mkdir, andrmto manage dependencies, generate media files from templates, and perform the browser automation tasks required for publishing.
Audit Metadata