Clawskillshield

# ClawSkillShield 🛡️

**Local-first security scanner for OpenClaw/ClawHub skills.**

## What It Does

- **Static analysis** for security risks and malware patterns
- **Detects**:
  - Hardcoded secrets (API keys, credentials, private keys)
  - Risky imports (`os`, `subprocess`, `socket`, `ctypes`)
  - Dangerous calls (`eval()`, `exec()`, `open()`)
  - Obfuscation (base64 blobs, suspicious encoding)
  - Hardcoded IPs
- **Risk scoring** (0–10) + detailed threat reports
- **Quarantine** high-risk skills automatically

## Dual-Use Design

- **CLI for humans**: Quick safety checks before installing skills
- **Agent API**: Importable functions for autonomous agents/Moltbots to proactively scan and quarantine risky skills (essential post-ClawHavoc)

## Quick Start

### CLI (Humans)
```bash
pip install -e .
clawskillshield scan-local /path/to/skill
clawskillshield quarantine /path/to/skill
```

### Python API (Agents)
```python
from clawskillshield import scan_local, quarantine

threats = scan_local("/path/to/skill")
if risk_score < 4:  # HIGH RISK
    quarantine("/path/to/skill")
```

## Zero Dependencies
Pure Python. No network calls. Runs entirely locally.

## Why This Matters
ClawHavoc demonstrated how easily malicious skills can slip into the ecosystem. ClawSkillShield provides a trusted, open-source defense layer—audit the code, run offline, stay safe.

---

**GitHub**: https://github.com/AbYousef739/clawskillshield  
**License**: MIT  
**Author**: Ab Yousef  
**Contact**: contact@clawskillshield.com

# ClawSkillShield  v0.1.1

**Local-first security scanner for OpenClaw/ClawHub skills  built for BOTH humans AND autonomous agents/Moltbots.**

Post-ClawHavoc essential: Static analysis to catch malware/risks before install.

## What It Detects

-  **Hardcoded secrets**  AWS keys, API credentials, private keys
-  **Dangerous calls**  `eval()`, `exec()`, `open()`, high-risk functions
-  **Risky imports**  `os`, `subprocess`, `socket`, `ctypes`
-  **Obfuscation**  Large base64 blobs, suspicious encoding
-  **Hardcoded IPs**  Potential exfiltration vectors
-  **Zero dependencies**  Pure Python, safe for agent/human use

**Why dual-use?** Humans get simple CLI safety checks; autonomous agents/Moltbots get importable functions for proactive self-defense.

## Installation

```bash
pip install -e .
```

## Quick Start (CLI)

```bash
# Scan a folder/skill
clawskillshield scan-local /path/to/skill

# Quarantine high-risk code
clawskillshield quarantine /path/to/skill
```

## Examples

###  Safe Scan (Low Risk)
Scanning a legitimate skill with normal file I/O:

```bash
$ clawskillshield scan-local .

============================================================
ClawSkillShield Report: .
============================================================

 MODERATE RISK - Review
Risk Score: 5.5/10.0

------------------------------------------------------------
Threats Found:
------------------------------------------------------------

 [WARNING] Risky import: os
  File: .\clawskillshield\analyzer.py
   Line: 4
   Direct OS access

 [WARNING] Dangerous call: open()
  File: .\clawskillshield\analyzer.py
   Line: 19
   Arbitrary execution or sensitive access

 [WARNING] Risky import: os
  File: .\clawskillshield\skill.py
   Line: 4
   Direct OS access

============================================================
```

###  High-Risk Detection
Scanning code with malware patterns:

```bash
$ clawskillshield scan-local ../test-risky

========================================================...