skills/mitsuhiko/agent-stuff/openscad/Gen Agent Trust Hub

openscad

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes the openscad binary via shell scripts to perform 3D rendering and model validation based on user-provided input files.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from .scad and .json files which may contain embedded instructions designed to influence the agent's behavior.
  • Ingestion points: The tools/extract-params.sh and tools/render-with-params.sh scripts read and parse content from external files.
  • Boundary markers: The instructions do not specify delimiters or safety warnings to ensure the agent ignores potentially malicious content within these data files.
  • Capability inventory: The skill possesses the ability to execute subprocesses and write rendered files to the local file system.
  • Sanitization: Input parsing is performed using regular expressions and standard JSON libraries without explicit sanitization for prompt injection patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:18 PM
Security Audit — agent-trust-hub — openscad