ClawHub
Back to skill

Security audit

Mobula - Smart Alerts & Monitoring

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent read-only crypto alert skill, with the main user risk being disclosed continuous monitoring and optional alerts through third-party messaging services.

Install only if you are comfortable giving the agent a Mobula API key and letting it store monitoring rules, wallet/watchlist identifiers, previous values, schedules, and alert destinations for ongoing heartbeat checks. Use dedicated or pseudonymous wallet addresses and notification channels when privacy matters, and confirm exactly what will be monitored and where alerts will be sent.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

YARA rule 'ransomware_behavior': Ransomware-like patterns (mass encryption, ransom notes) [malware]

Critical
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).

Content

Scanner excerpt · SKILL.md (reported line 211)May include surrounding context.

lume condition met 4. Only alert if BOTH conditions true

Other smart alert examples:

  • "Alert if ETH breaks $4K after consolidating $3,800-$3,950 for 3+ days"
  • "Alert if any of my watchlist tokens pump >20% with volume >5x average"
  • "Alert if BTC drops >3% but ETH is green (divergence signal)"

Pattern 5: Market Brief

Automated morning/evening market overviews.

Setup:

"Send market overview at 7am and 7pm on Telegram. Include BTC, ETH, SOL, and any top 100 token that moved more than 10%."

How it works:

Every scheduled time (7am, 7pm):

  1. Check major tokens (BTC, ETH, SOL, BNB) via mobula_market_multi
  2. Identify major moves (>5% in 24h)
  3. Check volume leaders
  4. Provide concise overview

Example brief:

text
📊 Crypto Market Brief - Feb 20, 7:00 AM

Majors:
- BTC: $67,234 (up 2.1% 24h)
- ETH: $3,456 (up 4.3% 24h)
- SOL: $123.45 (down 1.2% 24h)

Big Movers (24h):
- PEPE: up 23%
- ARB: up 15%
- AVAX: down 12%

Sentiment: Cautiously bullish. ETH lea

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The README explicitly says the agent will store monitoring conditions in memory and supports scheduled, autonomous checks using wallet addresses and alert destinations, but it does not warn users that these identifiers may persist across heartbeats, logs, or agent memory. Even though the skill is framed as read-only, persistent handling of wallet and routing data can expose sensitive financial profiling information, especially if memory, logs, or notification integrations are shared or compromised.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill description and surrounding onboarding do not prominently warn that portfolio, wallet, token, and alert content may be transmitted to third-party messaging services like Telegram, WhatsApp, or Discord. Users may unknowingly expose financial interests, wallet associations, or monitoring outputs to external providers, creating privacy and operational-security risk.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill’s activation guidance uses very broad phrases such as 'monitor', 'track', and 'watch', which can cause the agent to invoke this skill in contexts where the user did not clearly consent to ongoing monitoring or external alerting. In a skill that stores monitoring configs and can send alerts to Telegram/Discord/WhatsApp, overbroad triggers increase the risk of unintended data collection, persistence, and disclosure.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.