ClawHub
Back to skill

Security audit

FeedbackPulse

Security checks for vulnerabilities and agentic risk

Overview

The skill’s HR data access and survey drafting fit its stated purpose, with explicit approval for writes and cautions around sensitive information and software installation.

Connect only an authorized HR workspace and consider whether its employee information may be shared with your AI client. Prefer OAuth for read-only work and a native remote connection. If using the bridge, pin a reviewed version. Keep tokens out of chat, revoke unused credentials, and approve each write explicitly.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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)

MCP server referenced without pinned version: 'npx -y mcp-remote'.

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

The command npx -y mcp-remote can download and execute an unpinned package version without an installation confirmation, exposing the client to compromised or malicious future releases. This is a conditional supply-chain risk, not evidence that the package is currently malicious; exploitation could execute code with the client's permissions and expose locally accessible credentials or sensitive HR data. The legitimate service endpoint and user-approval rules do not mitigate execution of a compromised bridge package.

Content

No source excerpt is available for this finding.