The Agent Skills Directory

[SAFE]: The skill provides purely instructional content for topic evaluation. It does not perform any network operations, local file system access, or command execution.
[INDIRECT_PROMPT_INJECTION]: The skill processes external topic data which introduces a surface for indirect prompt injection. (1) Ingestion points: User-provided topics and outputs from a 'topic-generator' are processed in the SKILL.md context. (2) Boundary markers: No specific isolation markers or delimiters are defined to separate untrusted topic data from the skill's instructions. (3) Capability inventory: The skill has no access to dangerous tools, network functions, or shell commands. (4) Sanitization: No input sanitization or validation logic is present for the ingested content.

topic-reviewer