skills/raddue/crucible/mock-to-unity/Gen Agent Trust Hub

mock-to-unity

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data, specifically HTML mockups and visual specifications, to generate C# and USS code. This provides a surface for indirect prompt injection, where an attacker could embed instructions within a mockup to manipulate the agent's code generation output.
  • Ingestion points: Step 1 involves reading full source HTML files or describing images provided by the user.
  • Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are specified for the input data.
  • Capability inventory: The skill can write C# controllers and USS files to the Assets directory.
  • Sanitization: No sanitization or validation of the input content is mentioned before it is used to guide implementation.
  • [DATA_EXFILTRATION]: The skill instructions direct the agent to read and write to a file located at ~/.claude/skills/mockup-builder/references/theme-variables.md. Accessing and modifying hidden files within the user's home directory is a sensitive operation that can lead to data exposure or unauthorized configuration changes, particularly when the path is outside the standard project workspace.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 02:18 PM