eks-design

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to install and use the Mermaid CLI (npm install -g @mermaid-js/mermaid-cli && mmdc ...) for rendering architectural diagrams. It also involves the generation and execution of shell scripts (generate-docx.sh, generate-pptx.sh) when the user requests document exports.
  • [EXTERNAL_DOWNLOADS]: Fetches the mermaid-cli package from the official NPM registry. This is a well-known community utility used to support the skill's primary function of generating visual design artifacts.
  • [PROMPT_INJECTION]: The skill defines a workflow ('Stage 1: Input Assessment') that ingests untrusted external data in the form of requirements documents and technical notes.
  • Ingestion points: External requirements documents and meeting notes are processed in Stage 1 to extract technical context.
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to wrap the external input.
  • Capability inventory: The skill possesses capabilities for shell command execution (npm/mmdc) and file system operations.
  • Sanitization: There are no explicit instructions provided to the agent to sanitize or validate the content of the ingested requirements.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 11:00 AM
Security Audit — agent-trust-hub — eks-design