The Agent Skills Directory

[SAFE]: The skill contains legitimate administrative documentation and does not exhibit malicious behavior such as unauthorized data access or exfiltration. Metadata attribution to 'planetscale' appears to reflect the documentation source rather than malicious intent.
[COMMAND_EXECUTION]: The skill provides templates for administrative database operations.
Evidence: references/ps-cli-commands.md and references/partitioning.md contain commands for managing databases, branches, and partitions. These are accompanied by mandatory human-review warnings.
[EXTERNAL_DOWNLOADS]: The skill references resources from well-known and trusted providers.
Evidence: SKILL.md links to PlanetScale's official GitHub repository. references/monitoring.md suggests the standard pg_activity utility.
[PROMPT_INJECTION]: The skill's functionality involving the analysis of database schemas presents an indirect prompt injection surface.
Ingestion points: Database catalog queries and API-based schema retrieval as seen in references/index-optimization.md and references/ps-cli-api-insights.md.
Boundary markers: Absent; no specific instructions are given to ignore data-embedded instructions.
Capability inventory: Includes the ability to execute administrative commands and destructive SQL operations.
Sanitization: No sanitization of metadata is performed.

postgres