Back to skill
Skillv1.0.5
VirusTotal security
Website Usability Test Nova Act · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignApr 30, 2026, 3:59 AM
- Hash
- 87559a440eafd56ac027b2e70dc022ba5632b976c31cf0f8e8e2abda78dee6d7
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: nova-act-usability Version: 1.0.5 The OpenClaw AgentSkills skill bundle is designed for AI-orchestrated usability testing using Amazon Nova Act. It transparently discloses its data access (reading `~/.openclaw/config/nova-act.json` for an API key and writing local trace/report files) and explicitly warns about potential PII capture in trace files. Crucially, the skill implements robust safety guardrails, defining `MATERIAL_IMPACT_KEYWORDS` in `scripts/dynamic_exploration.py` and instructing the AI agent in `SKILL.md` and `references/nova-act-cookbook.md` to 'ALWAYS stop testing before actions that cause monetary impact, external communication, account creation, or data modification'. The Python scripts and markdown instructions consistently guide the AI agent to perform its stated purpose safely, without evidence of intentional harmful behavior, data exfiltration to unauthorized endpoints, or malicious prompt injection objectives. The use of external APIs (Nova Act, Anthropic Claude for persona generation) is aligned with its functionality.
- External report
- View on VirusTotal