The Agent Skills Directory

Indirect Prompt Injection (LOW): The skill is designed to ingest and analyze untrusted external project files, creating a vulnerability surface. 1. Ingestion points: Uses Read, Glob, and Grep tools to access external file contents. 2. Boundary markers: Absent; no instructions are provided to the agent to treat external code as untrusted data or use delimiters. 3. Capability inventory: The skill utilizes the Bash tool and references a Python execution command. 4. Sanitization: No evidence of input validation or content sanitization logic is present in the instructions.\n- Unverifiable Dependencies (LOW): The skill documentation recommends executing scripts/security_scan.py for automated validation. However, this script is not included in the skill files, making its behavior and potential for malicious command execution unverifiable.

vulnerability-scanner