The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the intervals-icu-pp-cli binary using go install from github.com/mvanhorn/printing-press-library and via npx from the @mvanhorn/printing-press-library package on NPM. These resources belong to the skill's author and are used to provide the primary functionality.
[COMMAND_EXECUTION]: The skill wraps shell commands by passing $ARGUMENTS directly to the intervals-icu-pp-cli binary. It uses the Bash tool to execute these commands.
[DATA_EXFILTRATION]: The wrapped CLI includes an output delivery feature (--deliver webhook:<url>) that allows the agent to POST command results to an external URL. While a documented feature of the tool, it represents a mechanism for data exfiltration if the agent is directed to use it maliciously.
[INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by processing external data from the intervals.icu REST API.
Ingestion points: Training data including activities, wellness history, and calendar events are pulled into the local environment via the sync and activity get commands (SKILL.md).
Boundary markers: None present; the skill does not explicitly instruct the agent to ignore instructions embedded within the fitness data.
Capability inventory: The skill can execute shell commands (Bash), write to local configuration files (~/.config/intervals-icu-pp-cli/config.toml), and perform network operations (API calls and webhooks).
Sanitization: No sanitization or filtering of the external API content is mentioned in the instructions.

pp-intervals-icu