The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The installation instructions suggest fetching the agentcli-helpers package from an external registry using the uv tool. This package provides the underlying notify command used by the skill.- [COMMAND_EXECUTION]: The skill relies on executing shell commands and piping output between processes. It specifically leverages platform-specific notification tools like osascript on macOS and notify-send on Linux.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it interpolates untrusted data from external sources into system notifications.
Ingestion points: The skill reads data from local files via cat and remote endpoints via curl.
Boundary markers: There are no markers or system instructions to prevent the agent from being influenced by instructions hidden in the data being notified.
Capability inventory: The skill possesses the ability to execute shell commands and display information to the user.
Sanitization: No input validation or sanitization is performed on the data before it is piped to the notification command.

desktop-notifications