The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands including curl to send notifications to a local endpoint (http://localhost:8888/notify) and cat to read state information from a local JSON file.
[EXTERNAL_DOWNLOADS]: Fetches data from several well-known security news sources including tldrsec.com, no.security, krebsonsecurity.com, thehackernews.com, schneier.com, and risky.biz.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted data from external websites. * Ingestion points: Headlines and summaries retrieved from external URLs. * Boundary markers: The instructions lack specific delimiters or instructions to the model to ignore embedded commands within the fetched content. * Capability inventory: The skill can execute shell commands (curl, cat) and manage local state files. * Sanitization: No sanitization or validation of the external content is performed before it is processed.

SECUpdates