mono-color

Warn

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The included scripts scripts/build_design_system_board.py and scripts/build_vibe_coding_poster.py utilize subprocess.run to call the external command-line utility rsvg-convert for SVG-to-PNG conversion. While the binary path is verified using shutil.which and arguments are structured, the runtime invocation of system binaries represents a dynamic execution pattern.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process arbitrary user-supplied themes, phrases, and article ideas to produce generative image prompts. Ingestion points: User-provided subject, text, and intent are captured in SKILL.md. Boundary markers: The skill uses a structured YAML 'Recipe Manifest' to resolve inputs before compilation. Capability inventory: The skill uses image-generation tools and writes files to the user's filesystem. Sanitization: The instructions direct the agent to preserve exact supplied text and user-supplied wording without explicit escaping or filtering for prompt injection vectors.
  • [COMMAND_EXECUTION]: The instructions in SKILL.md direct the agent to perform file system operations, specifically creating and writing to the path ~/Desktop/Claude skills/mono-color/ to store generated assets.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 3, 2026, 03:34 AM
Security Audit — agent-trust-hub — mono-color