Back to skill

Security audit

Suricata IDS Monitor

Security checks across malware telemetry and agentic risk

Overview

The skill can perform useful Suricata alert triage, but its local-only privacy claims conflict with documented Telegram reporting, credential use, memory writes, and daily scheduling.

Review before installing. Use local-only portions cautiously, avoid making security logs world-readable, do not enable Telegram delivery unless you are comfortable exporting IDS findings to a third party, and only add the cron job if you want recurring report transmission and local memory retention.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (9)

Description-Behavior Mismatch

High
Confidence
96% confidence
Finding
The manifest presents the skill as read-only log triage, but the body also appends reports to memory files and installs scheduled execution. That mismatch can cause users to approve a passive analysis skill that actually persists data and creates ongoing automated actions.

Description-Behavior Mismatch

Medium
Confidence
96% confidence
Finding
The manifest presents the skill as read-only log triage, but the body also appends reports to memory files and installs scheduled execution. That mismatch can cause users to approve a passive analysis skill that actually persists data and creates ongoing automated actions.

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The skill accesses Telegram bot credentials from environment variables despite being described as a local log triage tool. Introducing credential access expands the skill's privilege surface and creates unnecessary secret exposure risk if the feature is not explicitly declared and consented to.

Intent-Code Divergence

High
Confidence
99% confidence
Finding
The documentation explicitly says 'Reads your local Suricata log only' and 'No data leaves your machine,' yet later instructions send reports externally and write them to disk. This deceptive or inaccurate documentation increases the chance of unauthorized disclosure of security telemetry and undermines informed consent.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs sending a threat report to Telegram without a clear warning that sensitive IDS/IPS findings, internal IPs, and signature data will be transmitted to a third-party service. Security alert data can reveal internal topology, incidents, and defensive coverage, so silent external transmission is risky.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The scheduled job automates recurring Telegram transmission and local report storage without clearly warning users about continuous disclosure and persistence. This increases the blast radius because a one-time approval can become ongoing exfiltration of sensitive monitoring data.

External Transmission

Medium
Category
Data Exfiltration
Content
```python
# After building the report string above:
import os, requests
requests.post(
    f"https://api.telegram.org/bot{os.environ['TELEGRAM_BOT_TOKEN']}/sendMessage",
    json={"chat_id": os.environ['TELEGRAM_CHAT_ID'], "text": report},
    timeout=10
Confidence
97% confidence
Finding
requests.post( f"https://api.telegram.org/bot{os.environ['TELEGRAM_BOT_TOKEN']}/sendMessage", json=

External Transmission

Medium
Category
Data Exfiltration
Content
# After building the report string above:
import os, requests
requests.post(
    f"https://api.telegram.org/bot{os.environ['TELEGRAM_BOT_TOKEN']}/sendMessage",
    json={"chat_id": os.environ['TELEGRAM_CHAT_ID'], "text": report},
    timeout=10
)
Confidence
94% confidence
Finding
https://api.telegram.org/

Env Variable Harvesting

High
Category
Data Exfiltration
Content
# After building the report string above:
import os, requests
requests.post(
    f"https://api.telegram.org/bot{os.environ['TELEGRAM_BOT_TOKEN']}/sendMessage",
    json={"chat_id": os.environ['TELEGRAM_CHAT_ID'], "text": report},
    timeout=10
)
Confidence
90% confidence
Finding
os.environ['TELEGRAM_BOT_TOKEN']

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.