Security audit

Agent Rule Pilot

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward agent-config synchronization helper that can change multiple project config files, but that behavior is disclosed and aligned with its purpose.

Before installing or using it, expect it to read and potentially update AI-agent rule files in the current project. Review the displayed diffs carefully before accepting changes, because these files can influence future agent behavior across several tools.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

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 (4)

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The top-level description advertises easy syncing across many agents but does not warn that the action may write to multiple configuration files across the repository. This can mislead users about the operation's side effects, increasing the risk of accidental bulk modification, config drift introduction, or unintended overwrites.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The trigger phrase "sync my agent configs" can cause broad file-writing behavior across multiple agent configuration files, but it does not clearly communicate scope, targets, or confirmation boundaries in the invocation itself. In a skill that performs multi-file sync operations, overly simple activation increases the chance of accidental execution or invocation in the wrong repository/context.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The phrase "check config drift" is ambiguous because it does not specify which configs, repository, or comparison baseline should be used. While this is primarily a read/report action, ambiguity can still lead to incorrect routing, unintended inspection of unrelated files, or misleading results in multi-project environments.

Vague Triggers

Low

Confidence: 88% confidence
Finding: The phrase "create RULES.md" is concise but still ambiguous without surrounding constraints, because many repositories may use similarly named files or discuss creating files generally. The document does not specify required context or exclusions for when this command should map to this skill.

Static analysis

No suspicious patterns detected.