Security audit

Brand Identity Analyzer

Security checks across malware telemetry and agentic risk

Overview

This is a real brand-analysis skill, but it requires saving generated profiles and pushing them to GitHub without a clear review or approval gate.

Install only if you intend to send brand names and prompts to Gemini/Google Search and save generated JSON for reuse. Treat the GitHub push instructions as requiring manual approval: review the generated file, confirm the target repository and branch, and do not let an agent push to main automatically.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (6)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 89% confidence
Finding: The skill advertises behavior that uses environment variables and writes files, but it does not declare corresponding permissions or boundaries. This creates a transparency and governance gap: an agent or user may trigger filesystem writes or secret-dependent execution without an explicit capability contract, making misuse and unintended side effects harder to detect or control.

Context-Inappropriate Capability

Medium

Confidence: 96% confidence
Finding: The skill's stated purpose is brand analysis and profile generation, but it additionally instructs operators to modify a repository and push to GitHub after every new profile. That expands the skill from local content generation into external publication and codebase modification, which can expose sensitive or incorrect data, create unauthorized commits, or be abused as a delivery path for broader repo changes.

Vague Triggers

Medium

Confidence: 80% confidence
Finding: The trigger language is broad enough to match generic requests about brand data or creative workflows, increasing the chance the skill is invoked outside the user's intended context. In combination with file-writing and downstream publishing behavior, over-broad activation can cause unnecessary data generation, storage, or repository changes from loosely related prompts.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The documentation normalizes automatic local writes and mandatory GitHub publication without discussing privacy, repository visibility, accidental disclosure, or the permanence of commits. This is dangerous because generated profiles may contain sensitive business context, internal notes, or erroneous data that becomes broadly accessible once pushed, and users are not warned about that risk before publication.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The script sends the user-supplied brand name and model prompt to Google's Gemini API with Google Search grounding enabled, which causes external network transmission of user input and provider-retrieved content. In a local agent skill, this is a real privacy/transparency issue because users may assume analysis is local unless explicitly warned, and the skill does not obtain consent or clearly disclose that third-party services will receive the query.

Missing User Warnings

Low

Confidence: 81% confidence
Finding: The script writes JSON output to a user-provided path and can also auto-create directories and save into a default catalog path without a prominent overwrite/create warning. This is not code execution, but it can unexpectedly overwrite files or populate workspace directories, which is a meaningful safety issue for an agent skill that may be run on behalf of a user.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.