oathe / scan / reports / dangerous

Is ASantsSec/openclaw-security-policy-check safe?

No. ASantsSec/openclaw-security-policy-check is dangerous to install. Oathe's behavioral security audit gave the openclaw-security-policy-check skill by ASantsSec a trust score of 38/100 with 5 findings (1 critical, 2 high). Report updated

https://clawhub.ai/ASantsSec/openclaw-security-policy-check

38
DANGEROUS

Is ASantsSec/openclaw-security-policy-check safe to install?

This skill exhibits dangerous malicious behavior by masquerading as a legitimate OpenClaw security audit tool while actually attempting to access sensitive credential files including SSH keys, AWS credentials, and environment files. The pattern of unauthorized access to diverse credential files is consistent with credential harvesting malware.

What security issues were found in ASantsSec/openclaw-security-policy-check?

Category Scores

Prompt Injection 85/100 · 30%
Data Exfiltration 0/100 · 25%
Code Execution 20/100 · 20%
Clone Behavior 70/100 · 10%
Canary Integrity 10/100 · 10%
Behavioral Reasoning 0/100 · 5%

Findings (5)

CRITICAL Unauthorized Access to Sensitive Credential Files -100 ▶

The skill accessed multiple sensitive credential files that have no relation to its claimed OpenClaw security audit functionality. This includes environment files, SSH private keys, AWS credentials, Docker configuration, and GCP credentials.

Evidence: Filesystem monitoring shows OPEN/ACCESS events for: /home/oc-exec/.env, /home/oc-exec/.ssh/id_rsa, /home/oc-exec/.aws/credentials, /home/oc-exec/.npmrc, /home/oc-exec/.docker/config.json, /home/oc-exec/.config/gcloud/application_default_credentials.json

HIGH Masquerading as Security Tool -100 ▶

The skill presents itself as a legitimate OpenClaw security audit tool but exhibits behavior consistent with credential harvesting malware. The pattern of accessing diverse credential files suggests malicious intent rather than legitimate security auditing.

Evidence: Skill claims to audit 'OpenClaw gateway configuration' but accesses unrelated AWS, Docker, SSH, and other credential files

HIGH Honeypot File Access Detected -90 ▶

The skill accessed multiple honeypot/canary files designed to detect malicious behavior. While files were not modified, unauthorized access indicates suspicious activity.

Evidence: Accessed canary files: .env, .ssh/id_rsa, .aws/credentials, .npmrc, .docker/config.json, .config/gcloud/application_default_credentials.json

MEDIUM Opaque Executable Code -80 ▶

The skill contains executable JavaScript code (audit.cjs) that was not provided for analysis, making it impossible to verify the actual behavior versus claimed functionality.

Evidence: File present: scripts/audit.cjs (content not disclosed in audit evidence)

LOW External Command Execution Instructions -15 ▶

The skill instructs the agent to execute external commands including 'openclaw security audit --deep' which could potentially be misused.

Evidence: SKILL.md contains: 'node {baseDir}/scripts/audit.cjs' and '运行 openclaw security audit --deep'

Should I install ASantsSec/openclaw-security-policy-check?

Oathe's verdict for ASantsSec/openclaw-security-policy-check is DANGEROUS with a trust score of 38/100. Recommendation: Do not install.