Multi Ai Search Analysis

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill appears to match its stated purpose of querying multiple AI chat services, but users should be aware it uses browser automation, external AI providers, and persistent logged-in browser sessions.

This skill is reasonable for its stated purpose, but use it with non-sensitive topics unless you are comfortable sharing the prompt with DeepSeek, Qwen, Doubao, Kimi, Gemini, and any other configured services. Review the Python scripts and dependencies before running setup, and consider using a separate browser profile or dedicated AI accounts.

Findings (5)

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

Anything included in the analysis prompt may be shared with several external AI services, not just Claw/OpenClaw.

Why it was flagged

The skill intentionally sends the user's question to multiple third-party AI providers and combines their answers.

Skill content
用 DeepSeek、Qwen、豆包、Kimi、Gemini 分别问下 [问题],最后给我综合报告
Recommendation

Do not use this skill with confidential, regulated, or private information unless you are comfortable sending it to the selected AI providers.

What this means

The skill may act through your logged-in AI service accounts and may consume account quotas or create chat history on those services.

Why it was flagged

The configured AI platforms require login, so the automation operates through user-authenticated accounts.

Skill content
"loginRequired": true
Recommendation

Use a dedicated browser profile or dedicated AI accounts if possible, and review each provider's account/session behavior before running broad analyses.

What this means

Local session data for AI services may remain in the skill directory after use.

Why it was flagged

The browser is configured to use a persistent local profile, which can preserve login/session state for repeated use.

Skill content
"userDataDir": "./browser-profile"
Recommendation

Treat the browser-profile directory as sensitive; remove it or log out of providers when you no longer want the skill to retain sessions.

What this means

Users may need to run local setup commands that are not represented in the install metadata.

Why it was flagged

The documentation instructs users to install Python dependencies and a Playwright browser even though the registry has no install spec.

Skill content
pip install -r requirements.txt
playwright install msedge
Recommendation

Review requirements.txt and the scripts before running installation commands, especially because the skill source is listed as unknown.

What this means

If a visited site exploited the browser, reduced sandboxing could increase local risk.

Why it was flagged

The browser automation is configured with '--no-sandbox', which weakens Chromium's normal process isolation.

Skill content
"args": [
      "--disable-gpu",
      "--no-sandbox"
Recommendation

Remove '--no-sandbox' unless it is required by your environment, and run the skill in a contained environment if possible.