Back to skill

Security audit

VS Code Copilot Custom Agent Creator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a markdown guide for creating VS Code Copilot custom agent files, with no bundled executable code or hidden behavior.

Install only if you want help authoring VS Code Copilot custom agents. Review generated agent files before use, especially any `tools`, `agents`, `handoffs.send`, user-level install paths, or `hooks` command entries, and prefer the minimum permissions needed for the agent's role.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: vscode-agent-creator
description: Create VS Code Copilot custom Agent (.agent.md) files.
---

# VS Code Copilot Custom Agent Creator

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger guidance is broad enough that the skill may activate for loosely related prompts such as generic 'create agent' or 'handoff' requests, increasing the chance of unintended use. In a skill that generates persistent agent configuration files, accidental invocation can lead users toward creating powerful automation artifacts without deliberate scoping.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill documents sensitive capabilities such as tool grants, sub-agents, handoffs, and especially command hooks, but provides no safety warning or guardrails about least privilege, trust boundaries, or execution risk. This is dangerous because it can normalize creation of agent files that automatically invoke powerful tools or local commands, enabling unsafe automation to persist beyond the current session.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.