ClawHub

MiroFish Predict

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed MiroFish simulation helper, with expected Docker, npm, and LLM API key risks but no evidence of hidden or malicious behavior.

Install only if you trust the MiroFish CLI and its Docker image. Use a limited or budget-capped LLM API key, avoid sensitive prompts, start with low simulation rounds, and stop the Docker backend when finished.

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 (3)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill description includes broad trigger phrases such as '推演', '預測', '模擬', and '如果…會怎樣', which are common forms of ordinary user questions. This can cause the agent to invoke the skill too often in situations where the user did not intend to run a Docker-backed, token-intensive simulation, increasing the chance of unintended execution and resource use.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The markdown says the workflow will automatically start the Docker backend and notes that the first run may pull an image, but it does not present this as a prominent safety warning before execution. A user or agent may trigger the skill without understanding that it will start containers and download images, which are significant side effects with security and system-impact implications.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill documents multiple API key storage locations, including environment variables and a local file, but does not warn about credential exposure, file permissions, shell history leakage, or avoiding echoing secrets into logs. Because the skill requires an LLM API key and may be used by automation, missing credential-handling guidance raises the risk of accidental secret disclosure.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal