Kimi

Security checks across malware telemetry and agentic risk

Overview

This instruction-only Kimi API helper is coherent and disclosed, with expected external API use and optional local notes gated by user approval.

Install only if you are comfortable using a Moonshot API key and sending approved prompt content to Moonshot. Keep the key in environment variables, redact sensitive data before API calls, and review any saved ~/kimi/ approval or routing notes periodically.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The setup instructs the agent to offer activation on broad phrases like "mention Kimi, Moonshot, long-context analysis, or Kimi-based coding," which can cause the skill to trigger in conversations where the user did not explicitly request this capability. That creates a scope-expansion risk: the skill may collect preferences, influence routing decisions, or prompt for file writes in contexts that only loosely relate to Kimi, increasing the chance of unintended persistence or external-provider handling.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal