Back to skill
Skillv1.1.2
VirusTotal security
ARK — Adaptive Regulative Kriptos (V1 Declarative) · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignApr 30, 2026, 4:12 AM
- Hash
- 83aea110eb9f87bb04b07d7f05e1608d7e37e0d01ae446422b0bb7fb06416563
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: ark-v1-core Version: 1.1.2 The skill bundle defines a comprehensive 'constitutional framework' (ARK) for AI agents, aiming to regulate their reasoning, truthfulness, integrity, and coherence. The `SKILL.md` and `ARK.yaml` files contain extensive prompt-level instructions for the AI agent, primarily directing it to perform self-analysis, identify inconsistencies within its own rule set (`ARK.yaml`), propose modifications, and activate specific tracers/ILRs for self-audit. Crucially, these instructions explicitly mandate human approval for any proposed changes ('do not suggest new ILRs or TRKs until my approval', 'the user manually applies each proposed modification'). There is no evidence of intent to steal secrets, exfiltrate data, install backdoors, execute arbitrary external commands, or bypass user intent in a harmful way. The instructions are focused on the AI's internal cognitive regulation and self-improvement, with clear safeguards for human oversight, aligning with the stated purpose of the skill.
- External report
- View on VirusTotal