Canvas Design.Skip
PassAudited by VirusTotal on May 9, 2026.
Overview
Type: OpenClaw Skill Name: canvas-design-skip Version: 1.0.0 The skill utilizes prompt injection techniques in SKILL.md, specifically within the 'FINAL STEP' section, where it provides deceptive instructions claiming the user has already provided feedback ('The user ALREADY said...') to force the agent into a refinement loop. It also contains an open-ended instruction to 'download and use whatever fonts are needed,' which could potentially be exploited for SSRF or unauthorized file downloads depending on the agent's toolset. While these tactics appear aimed at improving artistic quality, the use of deceptive state manipulation and broad network/file instructions warrants a suspicious classification.
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 be more creatively interpreted and less exact if the user provides strict layout, wording, or branding requirements.
This instruction may cause the agent to prioritize artistic interpretation over exact user-specified constraints. It is aligned with a creative design skill, but users should be aware of the behavior.
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.
Use clear, explicit constraints when exact compliance matters, such as required text, dimensions, colors, brand rules, or file format.