AgentConstitution

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent blockchain governance helper, with the main caution that any action logs written on-chain may be public and permanent.

Install only if you trust the contract addresses and are comfortable using a dedicated testnet/private key for this skill. Do not put secrets, PII, credentials, internal prompts, or sensitive operational details in logAction descriptions because on-chain transaction data can be publicly visible and difficult or impossible to remove.

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: 95% confidence
Finding: The skill instructs users to submit `logAction(...)` transactions with a free-form `description` and other metadata on a public blockchain, but it never warns that transaction inputs and emitted data are publicly visible and effectively permanent. This can lead agents or operators to leak sensitive operational details, internal identifiers, or behavioral context that can be monitored, correlated, and abused by third parties.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal