Claude Legal Skill

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only contract review skill whose behavior matches its stated purpose, with privacy and optional external-tool cautions users should understand.

Install is reasonable for first-pass contract review, but treat contracts as sensitive. Use only documents you are authorized to submit, redact unnecessary confidential or personal data, prefer enterprise or no-training AI environments for privileged matters, verify any optional GitHub-installed redline tooling before running it, and have qualified counsel review important agreements.

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
89% confidence
Finding
The README encourages users to feed contracts into AI tooling and to generate downstream redline deliverables, but it does not warn that contracts may contain confidential, privileged, trade-secret, or personal data. In a legal-review skill, that omission is security-relevant because users may upload sensitive agreements to third-party models or companion tools without understanding retention, training, or privilege-waiver risks.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal