Avatar Helper

AdvisoryAudited by Static analysis on Apr 30, 2026.

Overview

No suspicious patterns detected.

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.

NoteHigh Confidence
ASI01: Agent Goal Hijack
What this means

The agent may pick an avatar that does not match the user's preference if the user expected to control the final choice.

Why it was flagged

The skill makes the agent's avatar choice authoritative over the user's preferences within this task.

Skill content
用户可以建议,但**最终决定权在龙虾**,不是用户
Recommendation

Install only if you want the agent to make an autonomous avatar choice; require explicit final approval before using the avatar anywhere important.

NoteHigh Confidence
ASI10: Rogue Agents
What this means

The agent may initiate an unsolicited avatar-selection conversation after installation.

Why it was flagged

The skill encourages proactive post-install behavior, though it describes asking the user before proceeding.

Skill content
安装 skill 后自动执行:
1. AI 主动发消息问用户
Recommendation

Use this skill only if proactive agent messages are acceptable; otherwise invoke avatar selection manually.

What this means

The skill may save or overwrite workspace/assets/avatar.jpg with an image from the selected URL.

Why it was flagged

The documentation includes a shell command to download an external image into the workspace; this is purpose-aligned but mutates a local file path.

Skill content
curl -L -o workspace/assets/avatar.jpg "图片URL"
Recommendation

Confirm the image source and destination path before downloading, especially if an existing avatar.jpg file should be preserved.