The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for executing the terra CLI to manage account configurations and perform administrative actions via the terminal.
[EXTERNAL_DOWNLOADS]: The skill documents the installation of the terra CLI tool through official channels, specifically Homebrew (tryterra/tap/terra) and NPM (@tryterra/cli). These resources are provided by the official vendor.
[INDIRECT_PROMPT_INJECTION]: The skill enables the agent to read external data from the Terra API, creating a pathway for untrusted external content to enter the agent's processing context. 1. Ingestion points: The terra events payload retrieve and terra data-api commands fetch data from external API endpoints. 2. Boundary markers: The skill instructions do not specify the use of delimiters or warnings to ignore instructions embedded in the retrieved data. 3. Capability inventory: The skill has the authority to execute bash commands, specifically the terra CLI tool. 4. Sanitization: There is no mention of sanitizing or validating the API response data before it is processed.
[CREDENTIALS_UNSAFE]: The skill includes instructions for retrieving and revealing environment API keys using the terra environments api-key retrieve --reveal command, which constitutes a credential exposure surface intended for administrative use.

terra-cli