iHub日报自动填写

Security checks across static analysis, malware telemetry, and agentic risk

Overview

The skill’s iHub daily-report automation is mostly purpose-aligned, but the bundled script can print the user’s iHub password to output/logs.

Only use this with an iHub account you control, verify the report details before submission, and avoid running the bundled script unless the password-printing behavior is fixed or you are sure execution logs are private and temporary.

Static analysis

No static analysis findings were reported for this release.

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.

What this means

Your iHub password could appear in terminal output, agent logs, or other captured execution records.

Why it was flagged

The script accepts the user's iHub password and then includes it in the returned object that is printed to stdout when the script is executed.

Skill content
return { username, password, content, name, ... }; ... console.log(JSON.stringify(result, null, 2));
Recommendation

Do not print or return raw passwords. Use a scoped credential mechanism, redact secrets from output, and avoid passing passwords on command lines.

What this means

If the wrong date, name, or content is supplied, the agent may submit an incorrect daily report.

Why it was flagged

The skill is designed to save and submit changes in an authenticated iHub workspace. This is consistent with the skill purpose, but it is still mutation of account/workplace data.

Skill content
13. 点击「已完成」保存内容
14. 点击「提交」按钮
15. 填写变更理由并确认提交
Recommendation

Verify the target name, date, and Markdown content before invoking the skill; ideally require a final confirmation before clicking submit.

What this means

Injected page scripts can interact with the authenticated iHub page while the session is active.

Why it was flagged

The skill explicitly uses browser-page JavaScript injection to set form content. This is disclosed and purpose-aligned, but users should understand that the agent will execute code in the logged-in page context.

Skill content
使用 evaluate API 注入 JavaScript 填入内容(触发 Vue 双向绑定)
Recommendation

Keep the injected logic narrowly limited to the report field and avoid using this skill on unrelated pages or with untrusted report content.