v1.0.1

交互式生成公众号文章

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 7:57 AM.

Analysis

This is a coherent instruction-only WeChat article writing helper, with minor user-awareness cautions about AI-authorship transparency, optional external image-prompt use, and document export.

GuidanceThis skill is safe to treat as a writing assistant. Before installing or using it, be mindful that it encourages naturalizing AI-generated text, so you should fact-check content and disclose AI assistance when appropriate. If you use the suggested Doubao prompts or export Word documents, review the content, metadata, and any private information before sharing externally.

Findings (3)

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.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Human-Agent Trust Exploitation

SeverityLowConfidenceHighStatusNote

SKILL.md

“去除AI写作痕迹：润色策略让文章读起来像真人写的”

The skill openly asks the agent to make generated writing appear more human-like. This is purpose-aligned for style polishing, but it can affect reader transparency if the final article is presented as wholly human-authored.

User impactReaders may not realize the article was AI-assisted unless the user chooses to disclose it.

RecommendationUse the polishing guidance for readability, but fact-check all details and consider appropriate AI-assistance disclosure for public or professional posts.

Tool Misuse and Exploitation

SeverityInfoConfidenceHighStatusNote

SKILL.md

“用户确认文章和配图方案后，可选择导出WORD文档” and “支持使用自定义WORD文档模板”

The skill describes optional user-confirmed document export and template use. This is aligned with the stated writing purpose and does not show unsafe execution, but it can create or use local document files.

User impactThe agent may produce a DOCX artifact or use a user-provided template if requested, so the resulting file should be reviewed before sharing.

RecommendationOnly export documents after confirming the content, use trusted templates, and verify the DOCX formatting and metadata before publication.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Insecure Inter-Agent Communication

SeverityInfoConfidenceHighStatusNote

SKILL.md

“您可以将上述提示词复制到豆包（doubao.com）生成对应图片”

The skill suggests manually copying generated image prompts to an external provider. This is disclosed and user-controlled, but the prompts may contain article themes or business details.

User impactIf the article topic or prompt includes private information, copying it to an external image service may disclose that information to the provider.

RecommendationReview prompts before pasting them into Doubao or any other external service, and remove confidential or unpublished details.