show-me
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to save files using a user-influenced description as a filename and then execute the
opencommand on that file. If the filename is not properly sanitized by the agent, this could lead to command injection (e.g., using semicolons or pipes in the description). - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied topics to generate visualizations and HTML artifacts, creating a vulnerability surface where malicious content could be reflected into local files.
- Ingestion points: Processes user-provided descriptions and topics to generate visual content and filenames in
SKILL.md. - Boundary markers: No explicit instructions for handling untrusted content or using boundary markers are present to prevent the agent from being misled by embedded instructions in the user's topic.
- Capability inventory: The skill possesses the ability to write HTML files to the local filesystem and execute the
opencommand. - Sanitization: There is no mention of sanitizing, escaping, or validating the user-provided input before it is interpolated into filenames or HTML content.
Audit Metadata