The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional content and code snippets for interacting with Redis Search.
Evidence: All documentation and examples in SKILL.md and the references/ directory use standard commands like FT.CREATE, FT.SEARCH, and FT.ALIASADD.
[EXTERNAL_DOWNLOADS]: The skill references external documentation and source code repositories.
Evidence: URLs point to redis.io and github.com/redis/agent-skills. As these are official domains and repositories for the author 'Redis', they are considered safe.
[INDIRECT_PROMPT_INJECTION]: The skill provides templates for building search queries that might process untrusted data.
Ingestion points: Filter parameters in FT.SEARCH queries and document fields in FT.CREATE (e.g., in references/query-optimization.md).
Boundary markers: None explicitly shown for user-provided input in the code snippets.
Capability inventory: Capability to execute search, aggregation, and index management commands via Redis clients.
Sanitization: The skill recommends using DIALECT 2 in references/dialect.md, which provides better handling of special characters and predictable parsing.

redis-query-engine