skills/simonrw/skills/notify-me/Gen Agent Trust Hub

notify-me

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a bash script to execute the osascript system utility. This is used to display native macOS notifications. The script correctly handles the input message as a positional argument rather than direct string interpolation, which helps prevent shell command injection.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The message variable in SKILL.md is populated with content provided by the user or agent.
  • Boundary markers: None explicitly defined in the prompt instructions, though the shell script uses double-quotes for the variable.
  • Capability inventory: The skill can execute shell commands through the environment's bash and osascript tools.
  • Sanitization: The skill uses standard shell argument passing (-- "$message") and AppleScript's argv handling, which is a secure way to pass untrusted strings to a script without them being interpreted as code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 02:25 PM
Security Audit — agent-trust-hub — notify-me