Security audit

finishing-a-developm

Security checks across malware telemetry and agentic risk

Overview

The skill is not clearly malicious, but it asks for broad file, command, credential, and network capabilities that are poorly scoped for a branch-finishing helper.

Review before installing. Only use this if you can constrain it to a specific repository and a clear branch-finishing workflow, and do not provide API keys or broad shell access unless the publisher documents exactly which services, commands, and files are involved.

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 Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 93% confidence
Finding: The skill is declared as a branch-finishing helper, but the body broadens its scope to generic data processing, API usage, file access, and command execution. This mismatch increases the chance that an orchestrator or user will grant the skill more trust and permissions than justified, enabling unintended operations under an innocuous label.

Context-Inappropriate Capability

Medium

Confidence: 92% confidence
Finding: A branch-management skill has no clearly stated need for generic API keys, outbound service access, or broad environment preparation, yet the documentation instructs operators to configure credentials and external connectivity. This creates unnecessary secret exposure and network-capable behavior that could be abused if the skill is invoked in a permissive environment.

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: The documentation explicitly includes command execution as part of system preparation and operation, but without a narrowly defined command set tied to branch-finishing tasks. In a skill that already has Bash access, broad command guidance can lead to arbitrary shell execution, destructive local changes, or misuse of user-provided parameters.

Vague Triggers

High

Confidence: 96% confidence
Finding: The trigger condition states the skill should activate whenever the user needs to perform 'related operations,' which is excessively vague for a skill with file, write, edit, and Bash tools. Broad activation criteria increase the risk of accidental invocation for unrelated tasks, leading to unexpected command execution, file modification, or network use.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill references API keys, file access, command execution, permissions, and network connectivity, but does not provide clear user-facing warnings about what sensitive actions may occur or what data may be exposed. This weakens informed consent and makes risky operations seem routine, increasing the chance of unsafe deployment.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.