org-chart-creator
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (HIGH): The skill is designed to process external content such as org chart patterns and visual documentation requests. It lacks boundary markers or sanitization logic to separate data from instructions. Ingestion points: User-provided org chart descriptions and diagram patterns. Boundary markers: None specified in the instructions. Capability inventory: Access to
Bash,Write,Edit,Read, andGreptools. Sanitization: None mentioned. This combination allows a malicious payload within a diagram description to potentially trigger arbitrary commands via the allowedBashtool.- [Command Execution] (MEDIUM): The skill explicitly requestsallowed-tools: BashandWrite. While potentially useful for generating SVG or Mermaid files, granting shell access to a skill that handles untrusted visual content descriptions is a violation of the principle of least privilege and significantly increases the risk of system compromise.
Recommendations
- AI detected serious security threats
Audit Metadata