The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local validation script and a CLI tool.
Evidence: Executes bash scripts/validate_params.sh for parameter checking.
Evidence: Utilizes the sports-skills command-line interface for fetching data.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from external RSS feeds and Google News.
Ingestion points: News content enters the agent context through the fetch_feed and fetch_items commands described in SKILL.md.
Boundary markers: Absent. The instructions do not specify any delimiters or warnings to the agent to ignore instructions embedded within the fetched news articles.
Capability inventory: The skill has the capability to execute shell commands (bash) and use a specific CLI tool (sports-skills).
Sanitization: Absent. There is no evidence of content sanitization or escaping of the fetched news text before it is presented to the agent.

sports-news