The Agent Skills Directory

[SAFE]: Analysis of the skill instructions and provided Python code reveals no malicious patterns such as prompt injection, data exfiltration, or unauthorized command execution.- [SAFE]: The skill references standard, well-known libraries including asyncpg, numpy, sentence-transformers, and elasticsearch for its search implementation templates.- [SAFE]: The provided code templates for PostgreSQL and Elasticsearch follow typical patterns for hybrid search. A best practice violation was noted in the PostgresHybridSearch class where SQL query keys are dynamically interpolated, which could lead to SQL injection if used with untrusted input keys; however, this is an implementation detail within a provided template rather than a malicious feature of the skill itself.

hybrid-search-implementation