Security audit

Sonarr

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward Sonarr integration that can add and remove shows, including optional media-file deletion when explicitly invoked.

Install only if you want an agent to manage your Sonarr library with your Sonarr API key. Verify the configured URL points to the intended Sonarr instance, protect the credential file, and approve remove or --delete-files commands only when you intentionally want the library entry or media files removed.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 88% confidence
Finding: The manifest says the skill searches for and adds TV shows, but the documented commands also support existence checks, configuration discovery, removal, and optional file deletion. This mismatch can mislead users and reviewers about the skill’s destructive capabilities, increasing the chance that a user grants trust to a skill that can remove content or expose configuration details.

Description-Behavior Mismatch

Medium

Confidence: 90% confidence
Finding: The skill description omits that it can remove shows and optionally delete files, which is a materially more dangerous capability than simple search/add behavior. Hidden destructive functionality undermines informed consent and can lead to accidental data loss if operators rely on the manifest summary instead of reading the full skill contents.

Description-Behavior Mismatch

High

Confidence: 98% confidence
Finding: The skill metadata says it searches and adds TV shows, but the script also implements a remove command that deletes series from Sonarr and can optionally delete associated files. This hidden destructive capability increases the chance that an agent or user invokes behavior they did not consent to, causing loss of media library entries or data.

Context-Inappropriate Capability

High

Confidence: 99% confidence
Finding: The --delete-files option performs destructive deletion of media files, which is materially more dangerous than the stated purpose of searching and adding TV shows. In an agent skill context, this creates a high-risk mismatch where destructive filesystem-side effects may be available without clear user awareness.

Missing User Warnings

Medium

Confidence: 74% confidence
Finding: The script silently reads Sonarr URL and API key material from a credentials file without any user-facing disclosure at runtime. In an agent setting, undisclosed credential use can surprise users and enables authenticated actions against an external service without clear consent boundaries.

Missing User Warnings

High

Confidence: 97% confidence
Finding: The remove path deletes series immediately, and with --delete-files it also removes associated files, without any confirmation or secondary check. This makes accidental or prompt-induced destructive actions much more likely, especially when the skill is operated by an automated agent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.