Security audit

systematic-debugging

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a broad automation/debugging template with command execution and credential guidance that is not clearly scoped for users.

Review this skill carefully before installing. It is not clearly limited to debugging: it may prompt an agent to run local commands, process files, and use API credentials or remote services without enough detail about scope, consent, or data handling. Install only if you are comfortable supervising each command and credentialed operation yourself.

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 (7)

Description-Behavior Mismatch

Medium

Confidence: 95% confidence
Finding: The skill is presented as a debugging utility but its documentation expands scope into generic automation, data processing, and workflow orchestration. Combined with the declared exec capability, this mismatch can cause the agent or user to invoke a far more powerful skill than intended, increasing the chance of unauthorized command execution or unsafe handling of unrelated tasks.

Context-Inappropriate Capability

Medium

Confidence: 93% confidence
Finding: The guide introduces API key setup, credential use, and remote API connection steps even though the stated purpose is local systematic debugging. This broadens the trust boundary to external services and may prompt users or agents to expose secrets or send debugging context off-host without a justified need or clear warning.

Context-Inappropriate Capability

Low

Confidence: 79% confidence
Finding: The documentation claims GitHub-sourced validation and network-dependent behavior without explaining why network access is needed for a debugging skill. Unjustified external validation can lead to unnecessary outbound requests, data leakage about code or environment, and confusion about provenance or trust.

Vague Triggers

High

Confidence: 97% confidence
Finding: The trigger language says to use the skill for any bug, test failure, unexpected behavior, and also broad development automation, making invocation criteria extremely broad. In a skill with exec access, overbroad triggering can cause the agent to select and run a powerful skill in situations where command execution or automation was not intended by the user.

Vague Triggers

High

Confidence: 96% confidence
Finding: The automatic trigger description states the skill should trigger whenever the user needs related operations, which is vague and self-referential. This ambiguity raises the likelihood of unintended auto-invocation and unsafe command execution in contexts outside the user’s actual request.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill declares the exec tool but the user-facing description around the capability does not clearly warn that commands may be executed on the host. Users may treat the skill as passive analysis while it actually has active execution capability, creating a consent and safety gap.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The quick-start section instructs users to configure API keys and access credentials but does not provide a clear warning about secret sensitivity, storage risks, or possible transmission to external services. This can normalize unsafe secret handling and cause accidental disclosure during debugging or automation workflows.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.