ClawHub
Back to skill

Security audit

Zyt ai creation

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a legitimate Chanjing AI generation helper, but it can send long-lived Chanjing credentials to an environment-configured API host and the package is missing several documented scripts.

Install only if you trust the publisher and need this Chanjing integration. Do not set CHANJING_API_BASE unless it is the official Chanjing API or a host you fully control, and avoid adding credentials until the missing command scripts are present and reviewable.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Tainted flow: 'req' from os.environ.get (line 64, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Content
headers={"Content-Type": "application/json"},
        method="POST",
    )
    with urllib.request.urlopen(req, timeout=30) as resp:
        return json.loads(resp.read().decode("utf-8"))
Confidence
92% confidence
Finding
The code sends app_id and secret_key to a network endpoint derived from the CHANJING_API_BASE environment variable. If that environment variable is attacker-controlled, the skill can be tricked into exfiltrating long-lived credentials to an arbitrary server over HTTPS, which is especially dangerous in an agent/runtime context where environment variables may be injected by the caller or host.

Missing User Warnings

Medium
Confidence
81% confidence
Finding
The reference explicitly documents a `download_result` capability that downloads `output_url` assets to the local system, but it does not warn that this writes files to disk, discuss destination/path safety, or mention user confirmation. In an agent skill, undocumented local file writes can surprise users and increase risk of unsafe persistence, overwriting files, or saving untrusted remote content to the host environment.

Static analysis

No suspicious patterns detected.