The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted text content from external sources which introduces an indirect prompt injection surface.
Ingestion points: Data enters the agent context via fetch_feed and fetch_items commands from third-party RSS URLs and Google News search results as described in SKILL.md and references/api-reference.md.
Boundary markers: Absent. There are no structured delimiters or instruction-isolation guidelines instructing the agent to ignore any embedded directives within the fetched headlines or summaries.
Capability inventory: The skill handles text retrieval and parameter validation using scripts/validate_params.sh. It does not possess advanced destructive capabilities such as file system writing, execution of arbitrary subprocesses, or state mutation.
Sanitization: Absent. The skill does not outline specific parsing restrictions or data validation rules to sanitize incoming HTML tags or natural language instructions embedded within news items.

sports-news