The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions to ingest and analyze untrusted system data which could contain malicious content designed to influence agent behavior.
Ingestion points: The skill reads system logs and process information using commands like journalctl, dmesg, and tail /var/log/syslog as described in SKILL.md.
Boundary markers: The instructions lack delimiters or safety warnings to ignore instructions embedded within the analyzed logs or command outputs.
Capability inventory: The skill utilizes administrative capabilities such as service management and shell execution through the bash-linux, server-management, and bash-pro skills mentioned throughout SKILL.md.
Sanitization: There is no mention of sanitizing or escaping the content of logs before processing.

linux-troubleshooting