Proof

Security checks across malware telemetry and agentic risk

Overview

This is a coherent ProofEditor integration, but it will send shared document content, access tokens, and presence/activity data to proofeditor.ai when used.

Install this only if you want the agent to use proofeditor.ai for collaborative documents. Treat Proof share links and tokens as sensitive, prefer bearer/header token use when possible, and avoid using the skill for private drafts unless you are comfortable storing and editing them in Proof.

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 (2)

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The skill directs the agent to immediately connect to a third-party service, read document state, and broadcast presence as soon as a shared URL is provided. That causes external data transmission and user/activity disclosure without an explicit confirmation or warning, which can leak sensitive document access patterns and metadata to proofeditor.ai.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill defaults to creating and using remote shared documents for collaborative work without first warning the user that document contents will be uploaded to external storage and made available through share links/tokens. This can expose drafts, plans, reports, or other sensitive material to a third-party platform unexpectedly.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal