Back to skill

Security audit

Receiving Code Review

Security checks for vulnerabilities and agentic risk

Overview

This skill is a code-review workflow guide with some malformed routing metadata, but no hidden destructive, credential-stealing, or exfiltration behavior was found.

Install only if you want an agent to be more deliberate when handling code-review feedback. Be aware that the metadata should be cleaned up because it may trigger less precisely than intended, but the reviewed artifact does not show malicious behavior.

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger engine states the skill activates whenever the user requests something that 'matches the description above,' but the description is malformed and broad. This can cause unintended invocation in adjacent contexts, leading the agent to apply this skill's behavioral constraints when they were not explicitly requested.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The frontmatter description is corrupted with graph syntax and vague invocation language, which makes matching behavior unpredictable and overly permissive. An agent may activate this skill on loosely related prompts, causing instruction bleed and incorrect workflow selection.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

High
Category
Not specified by scanner
Confidence
88% confidence
Finding

The frontmatter description is written in Indonesian without any indication that language selection depends on user preference. In systems that use metadata for routing or prompt construction, this can force a locale mismatch, degrade comprehension, and cause accidental invocation or incorrect behavior for users expecting another language.

Content

No source excerpt is available for this finding.