Security audit

Jarvis Ci Flake Hunter 01

Security checks for vulnerabilities and agentic risk

Overview

This is a simple Markdown workflow skill with no code execution, persistence, credential handling, or hidden data access, though its scope is broader than its flaky-test name suggests.

The main thing to consider is scope: this skill is safe-looking but generic. Install it if you want a reusable planning/checkpoint workflow; tighten the description and examples if you only want it invoked for flaky CI or nondeterministic test work.

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 (3)

Vague Triggers

Medium

Confidence: 96% confidence
Finding: The skill description is broad enough to be invoked for a wide range of general development tasks rather than a narrowly defined flaky-test workflow. Overbroad trigger language can cause inappropriate routing, where the agent applies this skill outside its intended scope and follows generic process guidance in contexts that may require different safeguards or domain-specific controls.

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The opening instruction, 'Deliver a concrete, reusable workflow for development tasks,' defines a generic operational playbook instead of a specialized skill trigger. This increases the chance the skill will be selected as a catch-all workflow engine, which can override more appropriate specialized skills and broaden the blast radius of any downstream mistakes.

Vague Triggers

Medium

Confidence: 98% confidence
Finding: The example prompts are framed as generic planning and execution requests, which teach the system to associate this skill with ordinary productivity tasks rather than flaky-test analysis. Example prompts strongly influence routing behavior, so broad examples can cause frequent misapplication of the skill in unrelated contexts.

Static analysis

No suspicious patterns detected.