ClawHub
Back to skill

Security audit

veterinary-appointment-reminders

Security checks for vulnerabilities and agentic risk

Overview

The bundled skill has limited local access, but its script checks input fields instead of producing the promised appointment reminders.

Expect a validation report rather than a ready-to-send reminder, and correct the template's field names before relying on it. Review appointment details and clinic-provided preparation instructions yourself. Use a dedicated output folder because an existing vet-reminder.md is overwritten. Treat the optional Windows app as a separate, unreviewed download.

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
  • 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 (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The declared central capability is reminder drafting, but the implementation only validates field presence and length. It neither composes a reminder nor reproduces the brief's content, and it has no dedicated location check. Local input reading and report writing are supporting implementation details, not separate security mismatches. No unrelated network access or other hidden capability is evident.

Content

No source excerpt is available for this finding.

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

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.

The skill produces a field-validation report rather than drafting an appointment reminder.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The manifest promises clear veterinary appointment reminders containing preparation instructions, location, and policy notes. The implementation outputs only field names, character counts, missing-field notices, and generic guardrails; it never includes the supplied appointment details or generates owner-facing reminder text.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.