ClawHub
Back to skill

Security audit

Price Monitor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed e-commerce price monitor; its file writes, notifications, and optional scheduled scans fit that purpose.

Before installing, confirm that you want the agent to visit the listed product URLs, save price and alert data locally, and create any recurring scan only after you explicitly request and review its schedule.

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
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Ae1

High
Category
analysis-evasion
Content
- Handles platform-specific page structures (see [PLATFORMS.md](references/PLATFORMS.md))
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

Ae1

High
Category
analysis-evasion
Content
- Handles platform-specific page structures (see [PLATFORMS.md](references/PLATFORMS.md))
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

YARA rule 'backdoor_persistence': Backdoor persistence with malicious payloads (shell commands, SSH key injection, hidden root users) [malware]

High
Category
YARA Match
Content
cted via Discord, alerts route there automatically.

## Scheduled Scanning

The agent should set up a recurring cron job when the user asks for
automatic scanning:

```
Every 6 hours: scan tracked products, check alerts, notify if triggered
Every 24 hours: generate and save a daily summary report
Weekly (Monday): generate weekly comparison report
```

### Setup
```bash
# Example — add to user's crontab or OpenClaw cron
openclaw cron add --name "price-scan" \
  --schedule '{"kind":"every","everyMs":21600000}' \
  --payload '{"kind":"agentTurn","message":"Run scheduled price scan on all tracked products. Check alerts and notify me of any changes."}'
```

## Price History Storage

### Default format: JSONL
```
scripts/price_history.jsonl
```
Each line is a JSON object:
```json
{
  "timestamp": "2026-04-30T10:00:00",
  "product_id": 1,
  "product_name": "iPhone 15 Pro",
  "price": 4899.00,
  "currency": "RM",
  "url": "https://shopee.com.my/...",
  "seller": "Apple Flagship Store",
  "in
Confidence
75% confidence
Finding
YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).

Vague Triggers

Medium
Confidence
87% confidence
Finding
The skill description contains broad, high-level trigger language such as 'Use when the user wants to' followed by multiple loosely scoped intents. This can cause the agent to invoke the skill for general competitor research, reporting, or alerting requests even when the user did not specifically ask to use this skill, increasing the chance of unintended web access and file writes.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The natural-language setup examples instruct the agent to accept URLs and begin monitoring based on conversational input, but they do not clearly constrain when activation should occur or require explicit confirmation of side effects. In practice, this ambiguity can lead to accidental enrollment of targets, scheduled scans, or persistent storage from casual user messages or quoted text.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The file instructs the agent to create a recurring cron job that persists beyond the current session, but it does not explicitly require clear user consent or warn that this creates an ongoing automated task in the user's workspace/system. In an agent skill context, persistence is security-relevant because it can continue consuming resources, performing network activity, and generating notifications after the original interaction is over.

Static analysis

No suspicious patterns detected.