The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the things-cli package globally via npm (npm install -g things-cli). This introduces a dependency on a third-party package from a public registry.
[COMMAND_EXECUTION]: The skill relies on the Bash tool to execute open commands for macOS URL schemes and uses the things CLI tool to read and write task data.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect injection as it processes untrusted user input (task titles, notes, and tags) and incorporates it into shell commands.
Ingestion points: User-supplied task names and details are passed as arguments to shell commands in SKILL.md.
Boundary markers: Absent; the skill relies on standard shell quoting or URL encoding without explicit delimiters to separate instructions from data.
Capability inventory: The skill uses the Bash tool to execute arbitrary task-related commands and launch the open utility.
Sanitization: The instructions recommend URL-encoding for parameter values in URL schemes, but they do not provide specific sanitization or escaping logic for the things-cli command-line arguments, which could lead to argument injection if input is not carefully handled by the agent.

things-mac