Back to skill

Security audit

cloudnap

Security checks for vulnerabilities and agentic risk

Overview

The skill is a straightforward CloudNap EC2 management helper, but it can change cloud instance state and delete automation schedules without consistently requiring confirmation.

Review this skill before installing if CloudNap controls production or cost-sensitive instances. It should be used only where the agent is expected to manage EC2 state, and users should require explicit confirmation before deleting schedules or changing automation.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
88% confidence
Finding

The DELETE endpoint takes a raw scheduleId and the skill does not require strong validation or confirmation that the selected ID matches the user's intended target. In a tool-using agent, this creates a parameter abuse risk where a mistaken resolution or prompt manipulation could delete the wrong schedule, disrupting automated start/stop controls for production instances.

Content

Scanner excerpt · SKILL.md (reported line 166)May include surrounding context.

md
Removes a schedule.

**Request:**
DELETE /api/v1/schedules/{scheduleId}
Headers:
X-API-Key: {CLOUDNAP_API_KEY}

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill allows deletion of schedules without an explicit confirmation or warning step, creating a risk of accidental or prompt-induced destructive changes to automation. In an infrastructure-management context, removing schedules can cause availability, cost, or operational policy violations if done mistakenly.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 199)May include surrounding context.

md
## Security Guidelines

1. **Never ask the user for their API key** — it is injected securely via `CLOUDNAP_API_KEY`.
2. **Never print or repeat the API key** in any response, log, or error message.
3. **Never read or access any local files or environment variables** other than `CLOUDNAP_API_KEY`.
4. **Only make outbound calls to** `https://cloudnap.in/api/v1` — no other domains.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The documentation explicitly requires a confirmation step before destructive stop operations. However, earlier action guidance says to use the stop endpoint when the user says commands like "stop" or "shut down," without including the required confirmation gate, creating contradictory operator instructions within the skill file.

Content

No source excerpt is available for this finding.