The Agent Skills Directory

[SAFE]: The skill provides instructional content for journalism quality control and news production workflows. No malicious patterns, data exfiltration attempts, or unauthorized command executions were found.
[PROMPT_INJECTION]: The 'AI FIRST' principle and iteration requirements are legitimate quality-assurance instructions designed to improve output quality; they do not attempt to bypass safety filters or override the agent's core instructions.
[DATA_EXFILTRATION]: References to external sites are limited to established journalism and intelligence resources (e.g., Reuters, AP, Poynter). No access to sensitive local files or unauthorized network activity was detected.
[REMOTE_CODE_EXECUTION]: No executable code, remote script downloads, or dynamic execution patterns are present. The skill references internal workflow templates and documentation stored on trusted platforms.
[INDIRECT_PROMPT_INJECTION]: The skill describes processing external documents (e.g., 'primary source verification', 'news gathering'). While this introduces a standard ingestion surface for news-related agents, the skill includes rigorous multi-level review protocols to mitigate accuracy and reliability risks.

editorial-standards