Security audit

Plausible Analytics

Security checks across malware telemetry and agentic risk

Overview

This Plausible Analytics connector is not malicious, but it under-labels an action that can write analytics events as if it were safe read-only access.

Install only if you intend to use OOMOL with a connected Plausible Analytics account. Treat `record_event` as a write action: confirm the exact site, event name, payload, and expected analytics impact before running it, even though the skill does not label it that way.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Description-Behavior Mismatch

Medium

Confidence: 91% confidence
Finding: The skill metadata says it should be used for 'searching and reading data', but the documented actions include `record_event`, which can write analytics data. This mismatch can mislead an agent or user into invoking a state-changing operation under the assumption the skill is read-only, increasing the risk of unauthorized or unintended analytics event injection.

Intent-Code Divergence

High

Confidence: 97% confidence
Finding: The safety section states that untagged actions are reads, but `record_event` is untagged even though it explicitly records pageviews or custom events. This creates a direct safety-policy contradiction that could cause an agent to execute a write action without confirmation, leading to false analytics data, event spam, or integrity issues in reporting.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.