ClawHub
Back to skill

Security audit

AgentAlarm

Security checks for vulnerabilities and agentic risk

Overview

The skill is a clearly disclosed iPhone alarm integration that stores a pairing token locally and sends encrypted alarm payloads through its relay.

Install only if you want an agent to ring your iPhone through AgentAlarm. Be aware it stores a local send token, uses the AgentAlarm relay by default, can schedule cron alarms when asked, and may ring through silent mode; review or pin AGENTALARM_RELAY if you do not want relay overrides.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (1)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill invokes a Node script and explicitly relies on environment variables and outbound network access, but it does not declare any tool scope such as permissions or allowed-tools. This creates an authorization gap: the runtime capabilities available to the skill are broader and less transparent than what the manifest communicates, which can lead to unintended network use, access to configuration secrets, or unsafe execution in environments that depend on manifest scoping for enforcement.

Content

No source excerpt is available for this finding.