ClawHub
Back to skill

Security audit

poe1-build-analyst

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent Path of Exile build-analysis helper that uses disclosed local files, public/user-approved data, and a constrained PoB CLI workflow.

Install only if you are comfortable giving the skill access to the PoB XML/code files and output paths you provide. Avoid passing private character data unless you intend to analyze it locally, and do not run public sharing commands unless you explicitly want a public link.

SkillSpector

Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill instructs the agent to perform network access, read and write files, inspect environment variables, and execute shell commands, yet it declares no explicit permissions boundary. That mismatch is dangerous because a host may expose these capabilities implicitly, allowing broader-than-expected data access or command execution from untrusted inputs such as PoB XML, character codes, or downloaded build data.

Description-Behavior Mismatch

Medium
Confidence
86% confidence
Finding
The script exposes an undocumented alternate execution path, `optimize-tree`, by importing and delegating to `poe1_tree_optimizer`. Hidden or out-of-scope modes increase attack surface because callers expecting only generate/compare/verify behavior may inadvertently permit more powerful operations from another module that is not analyzed here.

Static analysis

No suspicious patterns detected.