Security audit

Make Skill Template

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward template for creating GitHub Copilot Agent Skills and does not show hidden, destructive, or credential-seeking behavior.

Install this if you want a reusable guide for scaffolding Agent Skills. Review generated skills before enabling them, especially their discovery descriptions and any optional scripts they add.

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)

Ae1

High

Category: analysis-evasion
Content: 3. Update `SKILL.md`:
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The frontmatter description includes broad natural-language triggers like "create a skill" and "make a new skill," which can cause the skill to be auto-selected in a wide range of loosely related conversations. In a meta-skill that scaffolds new skills, overbroad discovery increases the chance of unintended invocation and propagation of weak patterns into newly generated skills.

Vague Triggers

Medium

Confidence: 85% confidence
Finding: The usage guidance lists several ambiguous activation conditions, such as wanting to add a specialized capability or duplicate the template, without defining scope boundaries or disqualifying cases. That makes accidental or inappropriate invocation more likely, especially because this skill influences creation of other skills and may spread permissive discovery language.

Vague Triggers

Low

Confidence: 80% confidence
Finding: The template explicitly encourages adding user prompt keywords for discovery but does not require precision, disambiguation, or negative examples. In a reusable template, this can normalize broad trigger design across many downstream skills, increasing ecosystem-wide prompt overmatching rather than causing direct code execution risk.

Static analysis

No suspicious patterns detected.