The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the leonardo-cli package from the NPM registry, which is the required interface for the Leonardo.Ai service.- [COMMAND_EXECUTION]: The skill utilizes shell command execution to interact with the leo CLI and uses curl for downloading generated assets, which is necessary for its stated functionality.- [DATA_EXFILTRATION]: The skill retrieves media assets from external URLs generated by the Leonardo.Ai API. These network operations are standard for the intended media generation workflow and do not involve unauthorized data movement.- [PROMPT_INJECTION]: The skill identifies an attack surface for indirect prompt injection via API responses and file metadata. * Ingestion points: Data is ingested through leo status --json and CLI output fields. * Boundary markers: The security policy in SKILL.md explicitly warns the agent to treat filenames and API content as data, not instructions. * Capability inventory: The skill has the capability to write files and execute shell commands via the CLI and curl. * Sanitization: The skill mandates the redaction of API keys if provided in user logs or prompts.

leonardo