Decision-Grade Reasoning (DGR)
Security checks across malware telemetry and agentic risk
Overview
This is a benign instruction-only reasoning aid, with the main caution that its outputs are intended to be stored and may summarize sensitive decisions.
This skill appears coherent and low-risk: it only guides the agent to produce structured JSON decision records. Use it as a documentation and review aid, not as proof that a decision is correct or compliant, and sanitize sensitive details before saving artifacts to tickets, incident systems, or audit logs.
VirusTotal
59/59 vendors flagged this skill as clean.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Sensitive decision context could become part of a longer-lived ticket or audit record with different access and retention controls.
The skill explicitly intends its generated decision artifact to be persisted outside the chat. This is disclosed and central to the audit purpose, but the artifact may contain sensitive business, incident, access-control, or financial decision details.
Store JSON artifact in ticket / incident / audit log
Before storing the JSON artifact, remove unnecessary secrets, personal data, or confidential details, and use appropriate access controls and retention policies for the destination system.