The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides numerous shell commands for system administration tasks, such as creating directories, writing configuration files, and managing Redis server processes. For example, examples/04-redis-cluster-setup.md contains instructions for setting up a 6-node cluster using mkdir, cat (for config writing), redis-server, and redis-cli.
[DYNAMIC_EXECUTION]: The skill documents and provides examples of Lua scripts for use with the Redis EVAL command to perform atomic operations. These are featured in SKILL.md (Section 5.2) and references/commands-admin-key.md (Section 3).
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection through the following components:
Ingestion points: The skill provides instructions and code examples for fetching data from external Redis servers using commands such as GET, HGETALL, LRANGE, and XREAD across multiple reference files.
Boundary markers: There are no instructions for the agent to use delimiters or ignore instructions embedded within the data retrieved from Redis.
Capability inventory: The skill includes powerful capabilities such as file system writes and process execution via redis-server and redis-cli (detailed in examples/04-redis-cluster-setup.md).
Sanitization: There is no evidence of input sanitization or escaping protocols for data retrieved from external stores before it is processed by the agent.

redis