Back to skill

Security audit

EchoTik-TikTok店铺商品列表

Security checks across malware telemetry and agentic risk

Overview

The skill mostly performs the advertised EchoTik seller-product lookup, but it has review-worthy privacy and storage issues.

Before installing, confirm you are comfortable using a LinkFox/EchoTik API key, spending credits, sending sellerId queries and some session/application metadata to LinkFox services, and storing full API responses locally. Pay particular attention to the automatic feedback API instruction and the mismatch between the documented storage path and the script's fallback to home or temporary directories.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (7)

Tainted flow: 'req' from os.environ.get (line 70, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Content
method="POST",
    )
    try:
        with urlopen(req, timeout=120) as response:
            return json.loads(response.read().decode("utf-8"))
    except HTTPError as e:
        body = e.read().decode("utf-8") if e.fp else ""
Confidence
94% confidence
Finding
with urlopen(req, timeout=120) as response:

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill instructs use of environment variables, outbound API access, and file writes, but no explicit permissions are declared. This creates a governance gap: the agent may access secrets, write persistent artifacts, and make network requests without transparent least-privilege review, increasing the chance of secret misuse, unintended data persistence, or unsafe external communication.

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The script always persists full API responses, cache entries, and session metadata to local storage, even though the skill's stated purpose is only to query seller product data. Persistent storage increases exposure of potentially sensitive business intelligence, account-linked metadata, and query history beyond the immediate operation.

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The docstring states output is restricted to <cwd>/linkfox and explicitly forbids /tmp, but the implementation silently falls back to home and temporary directories. This mismatch can cause sensitive response data to be written into broader or less-controlled locations than operators expect, undermining trust and storage policy assumptions.

Vague Triggers

High
Confidence
96% confidence
Finding
The trigger is explicitly broadened to fire even when the user does not mention EchoTik or TikTok, based only on loosely related store-product intent. Over-broad activation can cause unintended tool execution, unnecessary paid API calls, and data disclosure or side effects in contexts where the user did not consent to using this external skill.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The skill directs the agent to automatically send feedback to an external API whenever certain conditions occur, including user praise or dissatisfaction, and explicitly says not to interrupt the user's flow. This can exfiltrate user interaction metadata or content to a third party without notice or consent, creating a privacy and policy-compliance risk.

Missing User Warnings

Medium
Confidence
83% confidence
Finding
The script sends user-provided parameters together with session and application metadata to a remote service without an explicit notice at execution time. In a skill context, this is more sensitive because users may assume a local query helper rather than external transmission of identifiers and operational metadata.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.