show-math

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to use npx --yes mathshow, which fetches the package from the public NPM registry at runtime.
  • [REMOTE_CODE_EXECUTION]: By using npx --yes, the skill executes remote code from an external registry without requiring explicit user confirmation for the download.
  • [COMMAND_EXECUTION]: The skill relies on shell command execution (npx, node, mathshow) to perform its primary function. There is a risk of command injection if the agent interpolates user-provided LaTeX into the single-quoted shell command without proper escaping.
  • [INDIRECT_PROMPT_INJECTION]: The skill can be used to process external markdown files (e.g., mathshow notes.md), which may contain untrusted content.
  • Ingestion points: Local markdown files (like notes.md) provided to the CLI tool.
  • Boundary markers: None present in the instructions.
  • Capability inventory: Execution of shell commands and subprocesses via npx and node across all examples.
  • Sanitization: No sanitization or validation of the LaTeX input or file content is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 05:46 AM