The Agent Skills Directory

[SAFE] (SAFE): No malicious patterns detected. The skill contains only static markdown instructions and code examples for review purposes.- [Indirect Prompt Injection] (LOW): The skill is intended to process external data (source code from pull requests). While this is a surface for indirect injection, the skill lacks any capabilities—such as file writing, network access, or command execution—that could be leveraged by an attacker to cause harm. Evidence Chain: 1. Ingestion points: Input code snippets or PR data. 2. Boundary markers: Not defined in the skill body. 3. Capability inventory: None. 4. Sanitization: Relies on the host agent's native safety filters.

code-review