The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard administrative commands such as apt, dnf, systemctl, and redis-cli to manage system services. The script scripts/sk-redis-status.sh uses redis-cli to gather health and security metrics.
[PRIVILEGE_ESCALATION]: The skill requires elevated permissions to perform package installations and service restarts via sudo. These actions are within the expected scope of a Linux service management skill.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from command outputs, which presents a potential surface for indirect prompt injection.
Ingestion points: scripts/sk-redis-status.sh reads and parses output from redis-cli INFO and CONFIG GET commands. SKILL.md guides the agent to inspect journalctl logs.
Boundary markers: The script uses awk and tr to parse specific key-value pairs from command outputs, providing a degree of structural isolation.
Capability inventory: The skill possesses the ability to install packages, modify service configurations, and restart system units.
Sanitization: Data is filtered through awk to extract specific fields, reducing the risk of the agent interpreting raw command output as new instructions.

linux-inmemory-stores