The Agent Skills Directory

[SAFE]: No malicious patterns or executable code were found. The skill consists of markdown-based instructions and checklists for a persona.
[NO_CODE]: The skill does not provide any scripts, binaries, or automated tools, significantly reducing its attack surface.
[PROMPT_INJECTION]: The skill is designed to process external database code (migrations and queries), which is a surface for indirect prompt injection. However, because the skill has no functional capabilities like file writing or network access, the risk is negligible.
Ingestion points: User-provided code snippets analyzed against the provided checklists in SKILL.md.
Boundary markers: None.
Capability inventory: Restricted to generating text-based review findings; no subprocess, network, or file-write capabilities were identified.
Sanitization: None mentioned.

codereview-data