The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the ha command-line utility to perform task creation, list presets, and manage work hierarchies.
[INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface by instructing the agent to insert untrusted user input (such as task titles and work names) into shell command arguments.
Ingestion points: Arguments for --title in the ha task create and ha work create commands (SKILL.md).
Boundary markers: Absent; the instructions do not specify sanitization or delimiters for the user-provided titles.
Capability inventory: Shell command execution via ha task create, ha work create, ha work list, ha work show, ha agenda, ha task list, and ha preset list (SKILL.md).
Sanitization: Absent; no escaping or validation of the <title> or <work name> strings is described.

preset-trigger