A specialized security utility to compare different versions of Openclaw Skills and detect new, potentially dangerous capabilities before updating.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install arc-skill-differ
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install arc-skill-differ using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Skill Differ?
Skill Differ is a vital security component for developers managing Openclaw Skills. Unlike traditional scanners that look for known malicious patterns in a single file, this tool focuses on the evolution of a skill over time. It identifies when a skill gains new permissions or capabilities that were not present in previous versions, such as sudden network access requirements or file system modifications.
By analyzing the delta between versions, Skill Differ provides a layer of protection against supply chain attacks and unauthorized data exfiltration. It is particularly useful when updating Openclaw Skills sourced from community repositories or third-party marketplaces where code integrity must be verified manually.
Skill Differ Use Cases
- Comparing a currently installed version of Openclaw Skills against a new update from ClawHub.
- Auditing third-party Openclaw Skills for new credential access or network requests.
- Verifying that a version update only contains the bug fixes claimed by the developer.
- Performing quick security summaries before deploying agent updates in production environments.
How Skill Differ Works
- The user provides the paths to both the existing version and the new version of the Openclaw Skills directory.
- The differ engine crawls the file structure, specifically looking for security-relevant files and scripts.
- It compares code patterns to detect new capabilities such as network requests, file system access, or code execution (eval/exec).
- A recommendation is generated based on the severity of the changes, categorized as SAFE, REVIEW, or BLOCK.
- The tool outputs a detailed report or a JSON object for automated security pipelines.
Skill Differ Setup
The Skill Differ requires python3 and is typically located within your environment's scripts directory.
# Basic diff between an old and new version
python3 scripts/differ.py diff --old ~/.openclaw/skills/my-skill/ --new /tmp/my-skill-v2/
# Run with JSON output for automation
python3 scripts/differ.py diff --old ./v1/ --new ./v2/ --json
Skill Differ Data Schema & Taxonomy
The skill processes directory structures and outputs security classifications. Data is organized as follows:
| Attribute | Description |
|---|---|
| New Capabilities | Lists added permissions like net.access or sys.creds. |
| File Changes | Tracks additions, deletions, and modifications in scripts. |
| Recommendation | A status level (SAFE, REVIEW, BLOCK) based on detected risks. |
| Summary Mode | A high-level overview excluding specific line diffs for speed. |
Skill Differ Advanced Features
- Multi-format output including human-readable summaries and machine-parseable JSON.
- Granular capability detection for credential stealing and data exfiltration patterns.
- Deep inspection of script directories for newly added payload files.
- Heuristic detection of code obfuscation in updated Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
ARC Security MCP Server
A high-fidelity security intelligence service providing real-time auditing, static analysis, and semantic threat detection for the AI agent ecosystem.
Security Audit
A unified security assessment tool that chains scanners and verifiers to harden and secure your entire agent skill ecosystem.
Metrics Dashboard
A lightweight operational monitoring tool to track agent health, API usage, and performance metrics through a text-based dashboard.
Memory Pruner
A specialized utility to automatically prune, compact, and manage the size of AI agent memory and log files.
Agent GitOps
Automated deployment, rollback, and version management for AI agent workflows and skills.
Skill Health Monitor
A proactive monitoring utility designed to track performance drift, error rates, and resource usage for deployed AI skills.
METADATA
Featured*
