← Back to Scanner

openclaw Security Audit Report

🔭 Continuously monitored by ClawSecure Watchtower
Source:
SHA-256:

openclaw is an AI agent skill, created by openclaw and published at openclaw/openclaw. ClawSecure audited openclaw across 101 files through the 3-Layer Audit Protocol covering all ten OWASP ASI Top 10 categories, assigning a security score of 55/100 (Medium Risk). The 5 findings concentrate in Code Injection, Malicious Code and Permissions Manifest, including Attempts to access sensitive file: SOUL.md and Potentially dangerous code pattern detected: base64.*decode. 4 were rated high or critical severity.

Is openclaw safe?

ClawSecure audited openclaw and assigned a security score of 55/100 (Medium Risk), identifying 5 findings across Code Injection and Malicious Code. Review the findings below before installing.

What did ClawSecure find in openclaw?

ClawSecure identified 5 findings in openclaw, concentrated in Code Injection, Malicious Code and Permissions Manifest. 4 were rated high or critical severity. The most severe include Attempts to access sensitive file: SOUL.md and Potentially dangerous code pattern detected: base64.*decode.

How was openclaw audited?

ClawSecure ran openclaw through its 3-Layer Audit Protocol with full OWASP ASI Top 10 coverage, scanning 101 files from openclaw/openclaw.

What does a score of 55 mean?

ClawSecure assigned openclaw a security score of 55/100, placing it in the Medium Risk range. This reflects 5 findings led by Code Injection that warrant review before production use. ClawSecure derives this score with a weighted deduction model (critical -20, high -10, medium -5, low -2 from a base of 100).

Audit Findings for openclaw

ClawSecure detected 5 security findings in openclaw, spanning Code Injection, Malicious Code and Permissions Manifest.

Each finding is expandable in the interactive list below.

3-Layer Audit Protocol

Security Recommendations for openclaw

Eliminate dynamic code execution
openclaw evaluates code at runtime (for example eval or dynamic exec). Remove dynamic evaluation of untrusted input, and where code generation is unavoidable, sandbox it and validate every input.
Audit external network connections
openclaw connects to external endpoints. Verify every outbound connection goes to a trusted destination. Unauthorized callbacks are a primary indicator of ClawHavoc malware and data exfiltration. ClawSecure's proprietary engine monitors for known malicious endpoints including C2 infrastructure.
Add a config.json permissions manifest
A config.json file declares what an agent component can access: file system, network, shell execution and more. Without it, users have no visibility into what the component can do before installing. This is the single most impactful security improvement for any AI agent skill.
Pin dependencies to exact versions
Unpinned dependencies allow supply-chain attacks where a compromised version is pulled in automatically. Use exact version numbers in package.json (for example 1.2.3 instead of ^1.2.3) to keep unauthorized code out of your dependency tree. ClawSecure checks every dependency against known CVE databases.

Related Security Research

Why Generic Scanners Fail at AI Agent Security→Beyond Static Scans: Why ClawSecure Verifies Agentic Intent→ClawHavoc Explained: The Malware Family Targeting AI Agents→

Related AI Agent Security Audits

ClawSecScore 45/100keepScore 45/1004f12f0da6eb7a565Score 55/1007caf5d769dee89c3Score 55/100ClawSecScore 45/100

Scanned on July 22, 2026. openclaw is one of thousands of agents audited by ClawSecure from the community-curated awesome-openclaw-skills list and the openclaw/skills repository.

Start Your Free Scan