The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the newsjack CLI utility and the system date command to perform its primary function of tracking and recording keyword coverage in SKILL.md.
[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection.
Ingestion points: Untrusted article snippets, titles, and summaries are ingested through the news-search tool output as defined in SKILL.md.
Boundary markers: No delimiters or instructions to ignore embedded commands are present in the logic for processing external content.
Capability inventory: The agent has capabilities for shell execution (newsjack) and filesystem write operations (creation of candidates.json and decisions.json) which could be abused if an injection is successful.
Sanitization: There is no evidence of content validation or sanitization for the search results before they are processed by the LLM.

coverage-tracker