Security audit

Piv

Security checks across malware telemetry and agentic risk

Overview

This is a coherent software-development workflow skill that can modify repositories and run project commands, so it should be used deliberately but does not show malicious behavior.

Install only if you want an agent to actively work in a repository. Run it from the intended project, keep a clean git state or backup, review generated PRDs/PRPs and diffs, and approve shell commands, external research, and git commits deliberately.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Behavioral ASTexec() Call, eval() Call, Dynamic Import
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (7)

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill directs creation of directories and writing template/PRD/workflow files into the target project without any explicit user warning or confirmation. Because the skill is user-invocable and may default to the current working directory or auto-derived paths, it can unexpectedly modify repositories and leave persistent changes the user did not intend.

Missing User Warnings

Low

Confidence: 93% confidence
Finding: The skill explicitly instructs creation of `PRPs/planning/{feature-name}-analysis.md`, which is a repository-modifying action, but it does not warn the user that files will be written. In an agent setting, undisclosed write behavior can surprise users, cause unintended workspace changes, and be chained with broad invocation logic to modify repositories without clear consent.

Missing User Warnings

Low

Confidence: 90% confidence
Finding: The skill instructs the agent to write output directly to a user-supplied path (`$ARGUMENTS`) with a default filename, but it does not require confirmation before overwriting an existing file or warn about modifying files. In an orchestration skill that generates artifacts automatically, this can lead to accidental overwrites of important project documents or writes to unintended paths if arguments are mis-specified.

Missing User Warnings

Low

Confidence: 89% confidence
Finding: The skill explicitly instructs the agent to create a new markdown file in the repository during research, but it does not warn the user that repository contents may be modified. In an agent workflow, undisclosed write behavior can lead to unexpected state changes, polluted planning artifacts, or accidental commits, especially when the user expects analysis-only behavior.

Missing User Warnings

Low

Confidence: 95% confidence
Finding: The output section directs the agent to save the generated PRP to a specific path in the repository, but does not disclose this write action to the user. This is a real transparency and consent issue because the skill is an orchestrator and may be invoked in contexts where users do not expect filesystem modifications from a planning step.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill instructs the agent to automatically create directories and write a PRD file into the project workspace without any explicit user confirmation or warning that files will be modified. In an orchestration skill, this is risky because a user may believe they are only in a discovery conversation while the agent performs state-changing filesystem actions, which can lead to unintended workspace modification or pollution.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill explicitly instructs the agent to use write, edit, and exec capabilities to modify project files and run commands, but it provides no user-facing warning, confirmation gate, or safety boundary for those impactful actions. In an agent setting, this can lead to unintended code changes or execution of harmful project-supplied commands, especially because the executor is told to read PRPs and run validation loops that may themselves be adversarial or unsafe.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.