Security audit

Beckmann Knowledge Graph × Self-Improving Agent

Security checks across malware telemetry and agentic risk

Overview

This skill is a transparent workflow guide that coordinates two other skills, with the main caution being local learning logs that may persist task summaries.

Install only if you trust the two required skills and are comfortable with local .learnings files storing summaries of errors, insights, and Beckmann analyses. Avoid using it on sensitive personal, business, or strategic topics unless you plan to review or delete those logs.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill explicitly instructs persistent logging of errors and learnings to local files without any notice, consent gate, or data-minimization guidance. In practice, user prompts, task details, failures, and potentially sensitive content could be written to disk and retained beyond the session, creating privacy and confidentiality risk.

VirusTotal

56/56 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.