The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external news sources (Google News, Hacker News, and Reddit) to facilitate newsjacking workflows. While this ingestion of external data presents a potential surface for indirect prompt injection, the risk is mitigated by the skill's strict 'Send-approval handover' requirement, which mandates that every draft be reviewed and explicitly approved by the user before any external action is taken. \n- [COMMAND_EXECUTION]: The skill includes instructions for the agent to use standard CLI tools like curl, jq, and xmllint to fetch and parse news data from reputable platforms. These commands are used strictly for information gathering and do not facilitate the execution of untrusted scripts or remote code.

suede-public-relations