A specialized security skill designed to analyze script intent and detect malicious behavior before code execution.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install haoyuwang99-safe-exec
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install haoyuwang99-safe-exec using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
Safe Exec is a critical security utility within the Openclaw Skills ecosystem that empowers AI agents to vet code before execution. By performing a rigorous intent analysis, this skill identifies potential threats such as unauthorized network connections, data exfiltration, and obfuscated payloads. It moves beyond simple signature matching to provide deep reasoning about what a script actually does, ensuring that your environment remains secure from untrusted external code or third-party scripts.
To begin using Safe Exec, ensure your environment is configured for Openclaw Skills. You can trigger the analysis by passing an untrusted file path to your agent or using the integrated command interface.
# General installation for Openclaw Skills
openclaw install safe-exec
# Trigger manual analysis on a specific file
openclaw run safe-exec ./path/to/untrusted_script.py
Safe Exec organizes its findings into a clear, structured report format to provide transparency into the code's behavior:
| Attribute | Description |
|---|---|
| Intent Analysis | The name or description of the script or code block analyzed. |
| Verdict | A visual status indicator: SAFE, REVIEW, or BLOCK. |
| Summary | A concise 1-3 sentence plain-English explanation of the code's actual behavior. |
| Findings | A detailed list of specific suspicious triggers or techniques discovered during analysis. |
| Recommendation | Guided next steps to ensure secure execution within Openclaw Skills. |
Loading
CogDx provides external cognitive verification for AI agents to detect biases, audit reasoning, and ensure mathematical calibration of confidence levels.

A comprehensive computer vision tool for identifying vehicles, animals, and plants, alongside high-accuracy OCR for text and table extraction.

An automated presentation engine that generates self-contained, interactive HTML5 slidedecks using the Reveal.js framework via CDN.

A professional-grade financial tracking skill that reports upcoming or historical corporate events for a customizable watchlist of stock tickers.

A survival-focused cash flow decision system for solo founders that provides probability-weighted forecasting and runway alerts.

A technical framework for isolating and analyzing traffic originating from AI search engines and chatbots within standard analytics platforms.








































