The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides extensive examples of shell commands using redis-cli and standard Unix utilities like cat, grep, wc, and while loops for automation.
Evidence: Found throughout SKILL.md and reference files, such as scanning keys and inspecting types in a loop.
[PRIVILEGE_ESCALATION]: The installation guide includes commands that use sudo for system package managers.
Evidence: SKILL.md provides sudo apt install and sudo yum install instructions.
[DYNAMIC_EXECUTION]: The skill documents the use of Redis Lua scripting and the Redis Functions API for server-side logic.
Evidence: references/advanced-features.md contains syntax and examples for redis-cli --eval and FUNCTION LOAD.
[INDIRECT_PROMPT_INJECTION]: The skill outlines workflows that ingest data from Redis server outputs and external files, which constitutes a potential injection surface for the agent.
Ingestion points: redis-cli command output (including the MONITOR stream) and data piped from external files using -x or -X flags.
Boundary markers: The documentation suggests using standard shell quoting and explicitly mentions --raw versus --no-raw output modes.
Capability inventory: The skill utilizes file system writes via redirection, network communication via redis-cli host/port configuration, and server-side script execution.
Sanitization: Provides instructions for string quoting and escaping to handle binary-safe data.

redis-cli