Security audit

Performance Testing Toolkit

Security checks across malware telemetry and agentic risk

Overview

This appears to be a legitimate performance-testing skill, but it needs review because it encourages high-volume traffic to arbitrary URLs without clear authorization or safety guardrails.

Install only if you understand this is a traffic-generating testing tool. Use it only on systems you own or are explicitly authorized to test, start with low concurrency, avoid production credentials, and treat the reported stress-test results cautiously until the inconsistent CLI/docs and simulated stress-test path are fixed.

SkillSpector

By NVIDIA

Vulnerability Patterns

Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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 (10)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 86% confidence
Finding: The skill documentation clearly describes capabilities that require network access and likely file write access for report generation, yet no permissions are declared. This creates a transparency and governance gap: users or platforms may authorize or run a skill without understanding that it can send high-volume traffic to external systems and write output artifacts.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The README provides concrete load and stress testing commands that can generate substantial traffic against a target system, including a stress test up to 1000 concurrent users, without any warning about authorization, rate limits, production safety, or availability impact. In a performance-testing toolkit, this omission materially increases the chance that users will aim these commands at third-party or production systems and unintentionally cause service degradation or denial of service.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill teaches load and stress testing of target URLs without any safety guidance, authorization requirement, or warning about service disruption. Because this tool's intended function is to generate significant traffic, missing guardrails materially increases the risk of accidental denial-of-service against production or third-party systems.

Unpinned Dependencies

Low

Category: Supply Chain
Content: requests>=2.31.0 aiohttp>=3.9.0 asyncio dataclasses
Confidence: 92% confidence
Finding: Using `requests>=2.31.0` allows future versions to be installed without review and does not guarantee that all vulnerable ranges are excluded. In a tool that performs network-heavy testing, dependency drift can introduce regressions or newly disclosed supply-chain risk, making this more concerning than in a purely offline utility.

Unpinned Dependencies

Low

Category: Supply Chain
Content: requests>=2.31.0 aiohttp>=3.9.0 asyncio dataclasses pyyaml>=6.0.1
Confidence: 92% confidence
Finding: Using `aiohttp>=3.9.0` leaves the resolver free to install any later version, including versions later found vulnerable or behaviorally incompatible. Because this toolkit likely handles high-volume HTTP traffic, weaknesses in an async networking library can materially affect reliability and security.

Unpinned Dependencies

Low

Category: Supply Chain
Content: aiohttp>=3.9.0 asyncio dataclasses pyyaml>=6.0.1 jinja2>=3.1.2 psutil>=5.9.6 pytest>=7.4.0
Confidence: 95% confidence
Finding: `pyyaml>=6.0.1` is unpinned and PyYAML has a long history of unsafe deserialization issues when used incorrectly. In an enterprise testing toolkit, YAML is likely used for test configuration, so allowing uncontrolled version selection increases the chance of pulling in vulnerable or incompatible builds that could affect config parsing security.

Unpinned Dependencies

Low

Category: Supply Chain
Content: asyncio dataclasses pyyaml>=6.0.1 jinja2>=3.1.2 psutil>=5.9.6 pytest>=7.4.0 pytest-asyncio>=0.21.0
Confidence: 90% confidence
Finding: `jinja2>=3.1.2` is unpinned, so future releases or vulnerable intermediate versions may be resolved depending on environment and lock behavior. Since the toolkit generates reports, templating is a likely core function, which raises the importance of tightly controlling Jinja2 versions and usage.

Unpinned Dependencies

Low

Category: Supply Chain
Content: dataclasses pyyaml>=6.0.1 jinja2>=3.1.2 psutil>=5.9.6 pytest>=7.4.0 pytest-asyncio>=0.21.0 statistics
Confidence: 88% confidence
Finding: `psutil>=5.9.6` is not strictly pinned, allowing dependency drift into future versions without security review. For a performance-testing toolkit that likely inspects system metrics, flaws in a low-level process/system library can affect stability and potentially expose host information or crash conditions.

Unpinned Dependencies

Low

Category: Supply Chain
Content: pytest>=7.4.0 pytest-asyncio>=0.21.0 statistics matplotlib>=3.8.0 numpy>=1.24.0
Confidence: 86% confidence
Finding: `matplotlib>=3.8.0` is unpinned and may be involved in report generation, which makes reproducibility and security review important. Although less commonly associated with direct remote exploitation, unreviewed dependency drift can still introduce parsing or rendering-related issues.

Unpinned Dependencies

Low

Category: Supply Chain
Content: pytest-asyncio>=0.21.0 statistics matplotlib>=3.8.0 numpy>=1.24.0
Confidence: 90% confidence
Finding: `numpy>=1.24.0` is unpinned, permitting the installation of future releases or vulnerable ranges without explicit review. As a foundational native-code package, NumPy issues can have outsized effects on stability and security, especially where large datasets or file interactions are involved.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.