The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the alexacli tool from an external GitHub repository (github.com/buddyh/alexa-cli). This binary is not provided by a well-known or verified organization and is executed at runtime to interact with Alexa services.
[COMMAND_EXECUTION]: The skill's primary functionality is achieved through executing shell commands via the alexacli utility. This includes controlling smart home devices, playing media, and retrieving account data.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection by ingesting untrusted data from the Alexa API.
Ingestion points: The alexacli ask and alexacli history commands retrieve text responses and past voice activity from Alexa's servers, which are then processed by the agent.
Boundary markers: The instructions do not define specific delimiters or warnings to separate Alexa's output from the agent's system instructions.
Capability inventory: The skill possesses the capability to execute shell commands and interact with network services through the CLI tool.
Sanitization: There is no evidence of sanitization or filtering applied to the text returned by the Alexa API before it is consumed by the agent.

alexa-cli