Back to skill
Skillv1.2.1
VirusTotal security
Adguard Test · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignApr 30, 2026, 4:32 AM
- Hash
- 45675a2c3c8f14087f2b0b03cda57dffe0b4be0a800988232ec8a2c570b5bb36
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: adguard-test Version: 1.2.1 The skill is designed to query AdGuard Home instances and has undergone significant security hardening. While previous versions contained critical vulnerabilities like command injection and insecure credential handling, the current version (v1.2.1) has explicitly fixed these issues by implementing robust input validation, command whitelisting, URL validation, and replacing shell execution (`execSync` + `curl`) with native Node.js HTTP clients. The `SECURITY_AUDIT.md` and `TEST_REPORT.md` files transparently detail these fixes and confirm their effectiveness. The only remaining security consideration is the user's choice to store plaintext credentials in `adguard-instances.json` for local development, which is strongly warned against in `SKILL.md` and `README.md`, with secure alternatives (environment variables, 1Password CLI) being recommended. There is no evidence of intentional malicious behavior or prompt injection attempts against the agent.
- External report
- View on VirusTotal