← Back to Skills Marketplace
Ask Skill Shield
by
yx2601816404-sys
· GitHub ↗
· v1.0.0
340
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install ask-skill-shield
Description
When you encounter a technical problem, need a new capability, or want to find the right skill/tool for a task, ask the skill-shield agent before trying to s...
README (SKILL.md)
Ask Skill Shield
You don't need to browse 800+ skills yourself. Just ask.
When to Use This
- You need a capability you don't have (upload images, scrape web, send emails, etc.)
- You hit a technical error you can't solve
- You want to know if there's a skill/tool for something before building it yourself
- You're not sure if a skill is safe to use
How to Ask
Send a message to skill-shield via sessions_send:
sessions_send(
sessionKey="agent:skill-shield:main",
message="[skill-request] 我需要上传图片到抖音但一直失败，有没有相关的 skill 或解决方案？"
)
Message Format
Start with [skill-request] so skill-shield knows this is a skill consultation, then describe:
- What you're trying to do
- What's going wrong (if applicable)
- Any error messages
What You'll Get Back
- Matching skill names + how to install
- Security rating (A/B/C/D/F) so you know if it's safe
- Quick usage tips
- Alternative approaches if no skill exists
Examples
[skill-request] 我需要操作浏览器自动化，填表单、点按钮那种
[skill-request] 我想把 markdown 转成 PDF，有现成的 skill 吗？
[skill-request] 我在用 web_fetch 抓网页但拿到的是空的，JS 渲染的页面怎么办？
[skill-request] 有没有能帮我管理 cron 任务的 skill？
Why Ask Instead of Browsing
- skill-shield 已经扫描审计了 800+ 个 skill，知道哪些能用哪些有坑
- 很多 skill 名字和描述不直观，你搜不到不代表没有
- 有些需求不需要装 skill，skill-shield 可以直接告诉你怎么用现有工具解决
- 避免装到 F 级危险 skill
No Response?
If skill-shield doesn't reply within 60 seconds, it might be offline. Fall back to:
clawhub search "你的关键词"
Usage Guidance
This skill is coherent and low-risk: it only tells you how to ask a separate 'skill-shield' agent for recommendations. Before using it, avoid including passwords, API keys, or other sensitive data in your queries because they will be sent to another agent/service. Treat any safety ratings or 'audited' claims as advisory—if the skill-shield recommends installing a third-party skill, review that skill's permissions and metadata yourself before granting access.
Capability Analysis
Type: OpenClaw Skill
Name: ask-skill-shield
Version: 1.0.0
The skill 'ask-skill-shield' is designed to help the AI agent find appropriate skills, solve technical problems, or assess skill safety by consulting a specialized 'skill-shield' agent. The `SKILL.md` file provides instructions for the agent to use `sessions_send` to communicate with this helper agent. There is no evidence of malicious intent, data exfiltration, unauthorized execution, persistence mechanisms, or prompt injection attempts against the agent in the provided files. The instructions are clear, benign, and aim to improve the agent's decision-making and security posture.
Capability Assessment
Purpose & Capability
The name/description (find skill/tool recommendations and safety guidance) matches the SKILL.md content. No binaries, env vars, or installs are required, which is appropriate for a lookup/advice helper. The claim of auditing "800+ skills" is unverifiable from the package contents but is not inconsistent with the behavior.
Instruction Scope
Runtime instructions only explain how to format and send a query via sessions_send to sessionKey 'agent:skill-shield:main' and how to fall back to clawhub search. The instructions do not request files, secrets, or system paths. Note: using this skill will route your query to another agent/service — avoid including secrets or sensitive data in queries, since the skill's docs instruct sending free-text requests to an external agent.
Install Mechanism
No install specification or code files are present (instruction-only). This is the lowest-risk install model and matches the skill's stated purpose.
Credentials
No environment variables, credentials, or config paths are requested. The skill does not ask for unrelated secrets or broad access.
Persistence & Privilege
always is false and the skill does not request persistent privileges or modify other skills' configurations. Autonomous invocation is allowed by default but not excessive here.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ask-skill-shield - After installation, invoke the skill by name or use
/ask-skill-shield - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
v1.0.0: Lightweight skill that teaches agents to consult skill-shield for skill discovery. Instead of browsing 800+ skills, agents send a sessions_send message describing their need and get matched recommendations with security ratings.
Metadata
Frequently Asked Questions
What is Ask Skill Shield?
When you encounter a technical problem, need a new capability, or want to find the right skill/tool for a task, ask the skill-shield agent before trying to s... It is an AI Agent Skill for Claude Code / OpenClaw, with 340 downloads so far.
How do I install Ask Skill Shield?
Run "/install ask-skill-shield" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Ask Skill Shield free?
Yes, Ask Skill Shield is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Ask Skill Shield support?
Ask Skill Shield is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Ask Skill Shield?
It is built and maintained by yx2601816404-sys (@yx2601816404-sys); the current version is v1.0.0.
More Skills