show-me

Warn

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The instructions direct the agent to use Bash(open path/to/show-me-{description}.html) to display generated artifacts. The {description} placeholder is intended to be a summary of the topic. If the agent incorporates unsanitized user input into this filename, it could lead to command injection through shell metacharacters (e.g., ;, |, or &).
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process arbitrary user-supplied conversational topics to generate visual aids, which exposes a surface for indirect instructions.
  • Ingestion points: User-provided conversation topics (SKILL.md).
  • Boundary markers: None are present to separate untrusted user data from the instructions used to generate artifacts.
  • Capability inventory: The skill possesses the ability to write files to the local system (HTML artifacts) and execute shell commands (open).
  • Sanitization: No instructions are provided to sanitize the {description} variable or the user input before it is used in file operations or shell commands.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 4, 2026, 06:48 AM
Security Audit — agent-trust-hub — show-me