The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied code snippets to identify security vulnerabilities, which constitutes an ingestion point for untrusted data. 1. Ingestion points: The skill accepts code for analysis through the /security command or general requests for security reviews. 2. Boundary markers: The instructions do not specify the use of delimiters or special markers to isolate user-provided code from the system prompt instructions. 3. Capability inventory: The skill acts as an advisor and does not invoke tools for file system access, network communication, or shell command execution. 4. Sanitization: No sanitization or filtering of instructions that might be embedded in code comments or strings is specified in the skill prompts.

security-scanner