ClawHub
Back to skill

Security audit

stayforge-api

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent hotel-search and booking integration, but it can create, cancel, or modify reservations and transmit personal/payment-related data without mandatory confirmation safeguards.

Install only if you want an agent to use StayForge for hotel searches and booking management. Before allowing booking, cancellation, modification, API-key signup, or price alerts, require the agent to show the exact details, explain what data will be sent, disclose any cost/refund/cancellation impact, and get your explicit approval.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

High
Confidence
96% confidence
Finding
The booking creation instructions describe how to submit guest details and a payment method to create a reservation, but they do not require an explicit warning that the action may create a real booking, transmit personal data, and incur charges. In an autonomous or semi-autonomous agent context, this creates a material risk of unauthorized commitments, privacy exposure, and financial harm.

External Transmission

Medium
Category
Data Exfiltration
Content
### Step 2: Sign Up via API
```bash
curl -X POST https://stayforge.vosscg.com/v1/keys \
  -H "Content-Type: application/json" \
  -d '{"email":"user@example.com"}'
```
Confidence
84% confidence
Finding
The skill instructs transmission of a user's email address to an external third-party endpoint to create an API key. Although this is expected functionality, it is still an external transfer of personal data, and the skill does not include a clear consent/privacy notice or data-handling limitation before sending it.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The cancellation and modification flows provide direct API calls for changing bookings but omit clear warnings that these actions can be irreversible, can affect refunds, and can alter trip costs or availability. That is dangerous because an agent could make consequential changes without ensuring the user understands penalties, lost reservations, or additional charges.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The invocation text is overly broad and encourages use whenever users mention general travel or accommodation needs, without clearly limiting when the skill should act or requiring explicit user intent to search or book. In an agent setting, this can cause unintended activation and progression into booking-related flows that collect personal data or prepare financially consequential actions.

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:47