ClawHub
Back to skill

Security audit

Kraken.io

Security checks across malware telemetry and agentic risk

Overview

This Kraken.io skill mostly behaves like a normal connector wrapper, but it under-labels image upload/optimization as read-safe, so users should review it before installing.

Install only if you are comfortable with OOMOL/Kraken.io receiving image content for optimization and storing results in connector transit. Treat `optimize_image` as a data-processing action, not a read-only action, and confirm the exact image source and intended output before running it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
89% confidence
Finding
The manifest claims the skill is for 'searching and reading data,' but the documented `optimize_image` action performs content processing and stores output in connector transit. This mismatch can cause users or higher-level agents to invoke the skill under the assumption that it is read-only, reducing scrutiny and bypassing confirmation safeguards appropriate for non-read operations.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The safety section states that untagged actions are reads, but `optimize_image` is untagged even though it sends image content for processing and stores results in transit. This creates a misleading safety contract that could cause automated agents to execute a non-read action without user confirmation, exposing data or triggering unintended processing.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The trigger text says to use this skill for 'ANY Kraken.io request,' which is overly broad and can lead to unintended invocation in contexts where direct API use, stricter review, or a different workflow would be safer. Broad routing instructions increase the chance that an agent will over-delegate tasks to the skill without considering whether the requested operation involves processing user content or non-read side effects.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.