ClawHub
Back to skill

Security audit

makefile-review

Security checks across malware telemetry and agentic risk

Overview

This Makefile review skill is mostly coherent, but it can move from review into broad Makefile modification without clear separate user approval.

Install only if you want an agent to inspect Makefiles and possibly suggest or apply Makefile target changes. Before allowing any `--apply` or `--scope all` workflow, review the proposed Makefile edits and make sure they cannot alter CI, release, install, or cleanup behavior unexpectedly.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The module documents a generation phase that can create and apply new Makefile targets, moving the skill from passive review into active repository modification. In an agent setting, this scope expansion is risky because generated Makefile targets may introduce unsafe commands, alter build behavior, or execute later in CI/CD without the user expecting the reviewer skill to make changes.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger list contains broad terms like 'build', 'make', 'automation', and 'portability' that are likely to appear in many normal conversations. This can cause the skill to activate outside its intended scope, increasing the chance of unsolicited command suggestions or workspace inspection behavior in unrelated contexts.

Shadow Command Trigger

Medium
Category
Trigger Abuse
Confidence
92% confidence
Finding
Using 'build' as a trigger risks conflicting with a built-in command of the same name. This ambiguity can cause the wrong action to fire, leading to unintended behavior, confusion, or execution of a review workflow when the user intended a different system command.

Overly Broad Trigger

Low
Category
Trigger Abuse
Confidence
90% confidence
Finding
The trigger 'make' is an extremely common word in technical and non-technical contexts, so it is prone to accidental invocation. In this skill, that broad matching is more risky because the workflow includes shell commands and repository inspection steps, which could be suggested or initiated in contexts not actually requesting a Makefile audit.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.