The Agent Skills Directory

[SAFE]: The skill is a collection of reference materials and does not contain any malicious instructions or security risks. No obfuscation or prompt injection attempts were detected in the source documentation.
[NO_CODE]: The skill is composed strictly of Markdown files and does not include any functional code, scripts, or external dependencies.
[SAFE]: The skill ingests untrusted SQL queries and EXPLAIN plans for analysis. 1. Ingestion points: SQL query and EXPLAIN plan inputs defined in the SKILL.md workflow. 2. Boundary markers: The output format uses triple-backtick markdown blocks to encapsulate code. 3. Capability inventory: No operational capabilities, such as network access, file system modification, or subprocess execution, were identified. 4. Sanitization: No specific input sanitization or filtering is defined. This ingestion surface is considered safe due to the complete lack of exploitable capabilities within the skill.

sql-optimizer