The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill frequently invokes npx -y endurance-coach@latest to execute various subcommands. This pattern downloads and runs the latest version of the package from the NPM registry at runtime without version pinning, which introduces a supply chain risk.
[COMMAND_EXECUTION]: The skill provides the agent with a query command capable of executing arbitrary SQL against the local coach.db database. While intended for advanced analysis, this provides a powerful data access vector that could be exploited.
[EXTERNAL_DOWNLOADS]: Core functionality of the skill is dependent on fetching the endurance-coach package from the public NPM registry. This external dependency is used for data assessment, load management calculations, and HTML rendering.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection due to its processing of external training data from Strava.
Ingestion points: Athlete activity data (names, descriptions, and metadata) stored in the coach.db database, populated via Strava sync.
Boundary markers: None identified; the skill instructions focus on structured output (YAML/HTML) but do not define delimiters or warnings for ingested data.
Capability inventory: Includes shell command execution via npx and arbitrary SQL querying.
Sanitization: No data validation or sanitization steps are defined for the imported athlete activity history.

endurance-coach