ClawHub
Back to skill

Security audit

Match Loop by John Perry

Security checks for vulnerabilities and agentic risk

Overview

This skill provides a disclosed two-agent coding and visual QA workflow, with no hidden code execution, persistence, or unrelated data access found.

Install this if you want Codex to run a multi-agent build-and-review loop that may edit project files, start local previews, use browser automation, take screenshots, run tests, and install needed dependencies when appropriate. Review those actions as you would for any coding assistant workflow, especially in projects with credentials or live services.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The instruction "Do not default to Safari. Use Chrome unless explicitly told otherwise" hard-codes a tool preference in natural language. Under the policy-violation category, this is a natural-language constraint that removes user choice unless they explicitly override it.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.