The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the 'eightctl' binary via a Go module from 'github.com/steipete/eightctl/cmd/eightctl@latest'. As this repository is not affiliated with a trusted organization or well-known service, it represents an external dependency from an unknown source.
[CREDENTIALS_UNSAFE]: In SKILL.md, the documentation instructs users to provide sensitive credentials ('EIGHTCTL_EMAIL' and 'EIGHTCTL_PASSWORD') which are accessed by the tool from environment variables or a local configuration file at '~/.config/eightctl/config.yaml'.
[COMMAND_EXECUTION]: The skill relies on executing the 'eightctl' CLI tool to perform operations such as checking status, changing temperatures, and managing alarms. The installation process also involves executing 'go' build commands.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes output from an unofficial API (Eight Sleep pod status and configuration). Malicious data returned from the API could theoretically influence agent behavior, though the risk is localized to the data processed.
Ingestion points: Tool output from 'eightctl status', 'eightctl alarm list', and 'eightctl schedule list' (SKILL.md).
Boundary markers: Absent.
Capability inventory: Execution of 'eightctl' CLI with various arguments (SKILL.md).
Sanitization: Absent.

eightctl