The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute hyperai CLI commands on the host system to search, describe, and call various integrated tools (e.g., Gmail, Meta Ads). These commands are used to bridge marketing workflows from the agent to a terminal environment.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: User-provided data is interpolated into shell commands via flags (e.g., --to, --subject) and JSON blobs (e.g., --json '{...}') in SKILL.md.
Boundary markers: The skill encourages the use of the --json flag and shell quoting to delimit structured input from command parameters.
Capability inventory: The skill can execute a wide range of marketing and communication tools via hyperai call and manage authentication contexts with hyperai connections use.
Sanitization: The skill relies on the hyperai CLI's internal schema validation (verified via describe --parameters) rather than explicit instruction-based sanitization.

hyper-cli