ClawHub
Back to skill

Security audit

Yufluent Clawhub Publish Yufluentcn Ad Optimize

Security checks across malware telemetry and agentic risk

Overview

This ad-optimization skill is mostly coherent, but it automatically falls back to a broader Yufluent agent endpoint and sends an API key plus campaign data to a configurable network destination.

Review before installing. Use only a Yufluent API key you are comfortable sending to this service, avoid setting TOKENAPI_BASE_URL unless you fully trust the destination, and do not include confidential campaign, customer, or account data unless your organization permits sending it to Yufluent cloud.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (5)

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
91% confidence
Finding
return requests.post(url, **kwargs)

Lp3

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding
The skill advertises code-capable behavior that uses environment variables, local files such as .env, and outbound network access, but it does not declare explicit permissions in a first-class permissions model. That mismatch reduces transparency and can cause an agent or user to grant broader trust than intended, especially since the workflow requires sending user-supplied business data and an API key to a remote Yufluent service.

Description-Behavior Mismatch

Medium
Confidence
81% confidence
Finding
The fallback path expands this module from a narrow skill-specific API client into a generic agent-orchestration client by calling /agent/turn when /skills/{id}/run fails. That increases the reachable behavior surface beyond the advertised ads-optimization purpose, which can cause unintended actions, broader prompt exposure, and execution through a more powerful orchestration interface.

Context-Inappropriate Capability

Medium
Confidence
80% confidence
Finding
The constructed agent/turn body enables broader assistant capabilities than a simple ads-coaching client needs, including an orchestration entrypoint and web_fast mode. In a skill advertised as a focused optimization coach, this creates unnecessary authority and data exposure, especially if the backend agent can use web or other tools not implied by the skill manifest.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The script collects user-supplied campaign details such as message, product, market, metrics, and context, then sends them to a remote cloud service via run_skill(...). In this file there is no disclosure, consent prompt, masking, or warning that potentially sensitive business or marketing data will leave the local environment, creating a real privacy and data-handling risk.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.