show-me
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user conversation data to generate visual artifacts, including HTML files. This creates a surface for indirect prompt injection where malicious content could be embedded in the generated visuals.
- Ingestion points: The skill processes the "current topic of conversation" (SKILL.md).
- Boundary markers: None identified; the instructions do not specify delimiters or warnings for the generated content.
- Capability inventory: The skill generates HTML files and utilizes a bash command to open them.
- Sanitization: No explicit sanitization or escaping mechanisms are described for the conversation data interpolated into the HTML artifacts.
- [DYNAMIC_EXECUTION]: The skill generates and opens HTML files at runtime based on conversation context using a template pattern (
show-me-{description}.html). This represents dynamic execution of UI content created from untrusted input. - [COMMAND_EXECUTION]: The instructions include a pattern for the agent to execute a shell command to display generated files to the user (
Bash(open path/to/show-me-{description}.html)).
Audit Metadata