ClawHub

Kiza Negotiator

Security checks across malware telemetry and agentic risk

Overview

This skill is a review item because it asks for marketplace account authority and advertises autonomous bidding, negotiation, escrow handling, and deal completion without clear safety limits or reviewed implementation details.

Install only after verifying the actual kiza-nego implementation and understanding what marketplace account powers it receives. Use a dedicated or least-privilege account, keep auto-respond and auto-pilot disabled until tested, require manual approval for bids, counter-offers, escrow steps, and deal closure, set strict financial limits, and confirm how to stop the agent and delete logs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill advertises broad autonomous negotiation, bidding, and marketplace interaction capabilities without clearly defining when it may act, what actions require user confirmation, or what external systems it can bind the user to. In an agent context, this ambiguity can lead to unintended offers, commitments, or transaction steps being taken on the user's behalf, increasing the risk of financial loss or unauthorized commitments.

Missing User Warnings

High
Confidence
95% confidence
Finding
The skill promotes auto-pilot negotiation, auto-responding, deal closing, and market watching, but it does not provide a prominent warning that the agent may autonomously make marketplace decisions with financial or contractual consequences. In this context, the absence of a strong warning is dangerous because users may enable automation without understanding that the agent could accept terms, place bids, or advance transactions faster than the user can review.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal