agent execution market

• Outcome: Review
• Latest audit: May 29, 2026 at 9:24 PM UTC
• Version:
  1.0.0

Overview

This skill is documentation-only but points users to a global external npm install and describes autonomous financial and cross-chain execution without enough safety boundaries.

Review this carefully before installing. Test in an isolated environment, inspect the npm package and source repository, avoid production wallets or sensitive data, and require explicit confirmation plus low limits for any financial, cross-chain, or irreversible action.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (3)

Missing User Warnings

Medium

• Confidence: 86% confidence
• Finding: The README promotes high-impact financial, DeFi, and cross-chain actions without prominent warnings about volatility, slippage, bridge risk, smart-contract failure, fees, or irreversible loss. In an agent-execution marketplace, examples like yield optimization and token bridging can encourage users or downstream integrators to automate risky actions with insufficient human review.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The skill advertises key generation, cryptographic signing, intent submission, and DeFi/cross-chain usage, but provides no warning about private key handling, signing risk, fund exposure, or irreversible transactions. In this context, users may treat the skill as safe operational guidance and expose secrets or authorize real on-chain actions without understanding the consequences.

Missing User Warnings

Low

• Confidence: 87% confidence
• Finding: The skill promotes WebSocket, REST API, and third-party integration features without warning that intent contents, execution metadata, and event streams may traverse networked services. This can lead users to unintentionally transmit sensitive business logic, personal data, or operational details to external systems without appropriate consent, minimization, or transport/security controls.

VirusTotal

65/65 vendors flagged this skill as clean.