ClawHub

NEAR Agent Skills

Security checks across malware telemetry and agentic risk

Overview

This appears to be a read-only NEAR blockchain analytics and gas helper that uses expected public APIs without wallet access, persistence, or destructive behavior.

Install only if you are comfortable with this skill making outbound requests to NEAR RPC and NearBlocks, and avoid querying account IDs you consider sensitive. Keep the lockfile and review dependency changes on upgrade.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
82% confidence
Finding
The skill advertises multiple network-dependent capabilities such as real-time gas estimation and on-chain analytics, but the manifest does not declare corresponding permissions. This creates a permission-transparency gap: operators may approve or run the skill without understanding that it performs outbound network access, which weakens sandboxing, review, and trust controls.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The function sends a user-supplied NEAR account identifier to the third-party service api.nearblocks.io without any disclosure, consent flow, or indication that data leaves the primary NEAR RPC path. Even though account IDs are often public on-chain, transmitting them to an additional external analytics provider creates an avoidable privacy and metadata-leak risk, especially in an agent context where users may not expect off-platform lookups.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal