Back to skill

Security audit

UI Foundry

Security checks for vulnerabilities and agentic risk

Overview

The skill is rated suspicious with medium confidence because scanners found a high severity vulnerable dependency and flagged location-search capabilities that appear beyond its stated UI design purpose.

Consider avoiding installation until the maintainer explains the place-search integration and updates the affected dependency. Skillspector analyzed 27 of 31 components fully, with partial analysis of four; the results are incomplete and do not establish that the skill is safe.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • 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
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

The supplied code demonstrates a mapping/place-search integration rather than UI design, implementation, or inspection. No connection to the declared design workflow is shown, making this an undeclared capability rather than an evident supporting implementation detail. The snippet is a README usage example; the underlying service implementation and its precise network behavior are not supplied.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

The supplied code implements a location-search backend integration rather than visual design, UI components, or screen inspection. Although place search could support a particular interface, it is a distinct application capability not disclosed in the declared purpose. Authentication and rate limiting are supporting safeguards, not separate mismatches. The snippet wires up the service but does not show its internal implementation or establish that a network request occurs during initialization.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: picomatch==4.0.7 — 1 advisory(ies): CVE-2026-112366 (picomatch vulnerable to ReDoS through single-branch repeated extglobs)

High
Category
Supply Chain
Confidence
85% confidence
Finding

The lockfile pins picomatch 4.0.7, matching the affected version identified by the supplied advisory; malicious repeated-extglob patterns could cause excessive CPU consumption and denial of service if passed to the matcher. Here picomatch is a transitive dependency of Vite and tinyglobby, so the principal exposure is during development or builds rather than necessarily in the deployed UI, and exploitation requires attacker influence over glob patterns that is not demonstrated in this file.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: file_read, network.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.