Canvas Design Anthropic
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: canvas-design-anthropic Version: 1.0.0 The skill bundle is designed to guide an AI agent in creating visual art and design philosophies. While SKILL.md uses aggressive prompt engineering techniques—such as 'CRITICAL' directives and simulating prior user feedback to force a 'masterpiece' mindset—the instructions are entirely focused on aesthetic output (PDF/PNG/MD). There is no evidence of malicious intent, data exfiltration, or unauthorized system access; the reference to the './canvas-fonts' directory is consistent with the stated purpose of professional typography.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The generated design may prioritize artistic interpretation over precise instructions, brand requirements, or layout constraints.
This tells the agent to treat user input as a creative foundation rather than a strict constraint. That is consistent with an art-generation skill, but users with exact requirements should notice it.
Some subtle input or instructions by the user that should be taken into account, but used as a foundation; it should not constrain creative freedom.
If installing or using this skill, clearly mark any requirements that are non-negotiable and review the generated files before using them.
Users have less information about the publisher or upstream project, though the supplied artifacts do not show executable or sensitive behavior.
The skill has limited provenance information. Because it is instruction-only with no install spec or code files, this is a transparency note rather than a material security concern.
Source: unknown; Homepage: none
Prefer skills from known publishers when possible, and review the visible instructions before relying on outputs for important work.