Security audit

Streaming Buddy

Security checks across malware telemetry and agentic risk

Overview

This skill appears to do what it claims: provide streaming recommendations using TMDB while storing a local viewing profile.

Install only if you are comfortable with your streaming services, watch history, ratings, inferred preferences, cached TMDB responses, and TMDB API key being stored in the workspace and with search terms being sent to TMDB. Delete $WORKSPACE/memory/streaming-buddy/ to reset the local profile, and use a TMDB key intended for this skill.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 79% confidence
Finding: The skill advertises shell-based handler usage and requires `jq`/`curl`, but does not declare corresponding permissions or execution capabilities. This creates a transparency and governance gap: users or hosts may invoke a skill that can access the shell and network without clear permission boundaries, increasing the chance of unintended command execution or data access if the implementation is unsafe.

Vague Triggers

Medium

Confidence: 85% confidence
Finding: The description includes very broad triggers such as general discussion of Netflix, TV shows, or 'what should I watch', which can cause the skill to activate during ordinary conversation without clear user intent. In a skill that tracks history and learns preferences, accidental activation is more dangerous because it can lead to persistent profiling or external API calls when the user did not knowingly invoke the assistant.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The skill explicitly stores detailed viewing history, ratings, likes/dislikes, favorite actors/directors, and inferred taste preferences in persistent workspace files, but the description does not provide a clear privacy warning or consent model. This is sensitive behavioral profiling data, and users may not realize that casual entertainment interactions are being retained and used to build a long-term preference profile.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The search function sends raw user-entered titles/queries to TMDB over the network and the skill also builds later recommendations from persisted viewing history and preferences, yet the script provides no user-facing notice or consent mechanism about this third-party data sharing. In a personal preference-tracking assistant, that creates a real privacy risk because entertainment interests and usage patterns are behavioral data that users may not expect to leave the local workspace.

Natural-Language Policy Violations

Medium

Confidence: 84% confidence
Finding: The script defaults region and language to German/Germany and coerces unsupported languages back to 'de', which can silently override user expectations and influence content selection, provider availability, and recommendation behavior. This is primarily a privacy/UX issue rather than code execution, but it can mis-handle user context and cause unintended data processing under the wrong locale.

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: TMDB API requests hardcode `language=de-DE` even though the script reads a configurable language value earlier, so the user’s actual language choice is ignored for outbound requests. In this skill context, that can expose a mismatch between user intent and transmitted metadata and produce systematically incorrect localized results.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.