GitHub发布

Security checks across malware telemetry and agentic risk

Overview

The skill is coherent, but it can use your logged-in GitHub account to create a public repository and publish local files without clear confirmation safeguards.

Install or use this only if you want the agent to publish files to GitHub through your logged-in browser. Make sure it asks before creating a repository, choose Public only intentionally, and review all files for secrets or private data before committing.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

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.

#
ASI02: Tool Misuse and Exploitation
Medium
What this means

Files could be published publicly to GitHub, potentially exposing private code, secrets, or other local content if the wrong files are selected.

Why it was flagged

The browser workflow directs creation of a public GitHub repository and committing selected local files, but does not require a final user approval or sensitive-file check before public publication.

Skill content
选择Public ... 选择本地文件或拖拽文件 ... 点击Commit changes
Recommendation

Before using it, require the agent to confirm the repository name, account, visibility, exact files, and commit message, and scan for secrets before clicking Commit changes.

#
ASI03: Identity and Privilege Abuse
Medium
What this means

Actions may be performed under the user's GitHub identity, including creating public repositories and publishing commits.

Why it was flagged

The skill relies on the user's existing logged-in GitHub browser session to create repositories and commit files, which is delegated account authority.

Skill content
浏览器已登录GitHub账号即可
Recommendation

Use only in a browser session for the intended GitHub account, and require explicit approval before any repository creation or commit.