ClawHub
Back to skill

Security audit

Yufluent Clawhub Publish Yufluentcn Ecommerce Render

Security checks across malware telemetry and agentic risk

Overview

This is a cloud rendering skill that uses an API key as disclosed, with some endpoint-scoping caveats users should understand before use.

Install only if you trust Yufluent with the product data you render and with the TOKENAPI_KEY account. Set TOKENAPI_BASE_URL only to a trusted endpoint, avoid untrusted or unexpected local services, and be aware that fallback agent routing exists even though normal use is the render endpoint.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Tainted flow: 'url' from os.getenv (line 271, credential/environment) → requests.post (network output)

Critical
Category
Data Flow
Content
}
    if body is not None:
        kwargs["json"] = body
    return requests.post(url, **kwargs)


def _raise_for_status(resp: requests.Response) -> None:
Confidence
94% confidence
Finding
return requests.post(url, **kwargs)

Description-Behavior Mismatch

Medium
Confidence
84% confidence
Finding
This rendering skill exposes generic agent orchestration endpoints (/agent/turn and /agent/outcomes) in addition to the expected rendering run path. That expands the capability boundary of the skill beyond its declared purpose, enabling a caller to trigger broader agent behavior or route data through orchestration APIs that may have different permissions, logging, or side effects. In a narrowly scoped ecommerce rendering skill, this mismatch increases risk because users and reviewers may assume the component only renders templates.

Context-Inappropriate Capability

Medium
Confidence
87% confidence
Finding
The code can submit outcome telemetry to /agent/outcomes even though the skill is described as an ecommerce rendering client. This creates an extra outbound data path that may transmit user content, identifiers, or execution metadata to the service without being necessary for image rendering, increasing privacy and data-governance risk. In this context, undocumented telemetry is more concerning because it is outside the user's likely expectations for a rendering tool.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.