diagram-design-editorial

Fail

Audited by Gen Agent Trust Hub on Apr 19, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The installation instructions direct users to clone a repository from an unverified GitHub account (github.com/cathrynlavery/diagram-design). Since the skill is loaded and executed from this untrusted third-party source, it constitutes a high-risk supply chain vector.
  • [COMMAND_EXECUTION]: The documentation requires users to execute several shell commands, including git clone, ln -s for symlinking, and open for local asset viewing, which involves direct interaction with the host system.
  • [PROMPT_INJECTION]: The brand onboarding feature fetches HTML from an arbitrary URL supplied by the user to extract style tokens, creating a surface for indirect prompt injection. Ingestion points: Website content fetched via the onboard command. Boundary markers: No explicit delimiters or instructions to ignore embedded instructions are present in the processing logic. Capability inventory: The agent can fetch external URLs and write tokens to local files (references/style-guide.md). Sanitization: No sanitization or validation of the external HTML content is specified before the agent processes it for styling information.
  • [SAFE]: The skill references font delivery from fonts.bunny.net, which is a well-known and recognized service for typography resources.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 19, 2026, 01:41 AM