The Agent Skills Directory

[PROMPT_INJECTION]: The skill functions by analyzing untrusted code provided by users, which creates a surface for potential indirect prompt injection attacks. 1. Ingestion points: User-supplied code snippets as defined in the SKILL.md instructions. 2. Boundary markers: Absent. The instructions do not explicitly direct the agent to ignore instructions that might be embedded in code comments. 3. Capability inventory: None. The skill does not utilize any tools or commands that allow for file system access, network communication, or shell execution. 4. Sanitization: Absent. The skill processes the input code directly to provide explanations.

code-explainer