The Agent Skills Directory

[SAFE]: The skill provides deep technical expertise for Redis 8.10, including data structure selection, persistence policies, and cluster management. No malicious instructions or hidden behaviors were detected.
[EXTERNAL_DOWNLOADS]: The skill references several external resources in NOTICE.md and SOURCES.yaml (e.g., official Redis and GitHub repositories) for attribution purposes. These references target well-known organizations and official software documentation, following standard research practices.
[COMMAND_EXECUTION]: The skill documents the use of standard Redis CLI tools and Docker commands for environment reproduction and diagnosis. These are legitimate administrative operations within the context of the skill's primary purpose.
[CREDENTIALS_SAFE]: The skill provides guidance on security hardening, specifically advising on the use of ACLs and the rotation of passwords. It correctly uses placeholders like {{password}} in examples and warns against hardcoding secrets in command logs or history.
[PROMPT_INJECTION]: No evidence of prompt injection or instructions to override agent behavior was found. The instructions are focused entirely on Redis expertise and best practices.
[OBFUSCATION]: The content is presented in clear text markdown and Python. No Base64-encoded strings, zero-width characters, or homoglyph-based URL obfuscations were detected.
[INDIRECT_PROMPT_INJECTION]: The skill includes evaluation scenarios (e.g., analyzing a cache service) that involve processing Redis keys and database row data. However, the skill explicitly provides mitigations, such as warning against building keys from unvalidated user input, to prevent common injection vulnerabilities.

redis