Know what you're installing before you install it.

Static security analysis for OpenClaw AI agent skills. Detects prompt injection, credential theft, obfuscated payloads, and dangerous capability combinations.

19,461

Skills scanned

62.4

Avg trust score

3,183

Dangerous skills (16.4%)

6,182

Trusted skills (31.8%)

Trust Distribution

Trusted — 6,182 (31.8%)

Caution — 6,092 (31.3%)

Risky — 4,004 (20.6%)

Dangerous — 3,183 (16.4%)

Key Findings

4,895

Critical findings across all skills

1,457

High severity findings

10,562

Total findings detected

Most Common Capabilities

file_write

6,186 (31.8%)

credential_access

5,953 (30.6%)

network_out

4,081 (21%)

network_in

2,703 (13.9%)

package_install

2,666 (13.7%)

file_read

2,278 (11.7%)

data_encoding

1,077 (5.5%)

agent_memory

674 (3.5%)

process_exec

273 (1.4%)

dynamic_eval

146 (0.8%)

Browse Full Registry

Or read the State of OpenClaw Security report.