ClawHub
Back to skill

Security audit

Unified LLM Gateway - One API for 70+ AI models. Route to GPT, Claude, Gemini, Qwen, Deepseek, Grok and more

Security checks across malware telemetry and agentic risk

Overview

This is a coherent LLM routing skill that sends user-selected prompts or image references to the disclosed AIsa API using a user-provided key.

Install only if you intend to use AIsa as a third-party LLM gateway. Use a dedicated, revocable AISA_API_KEY, monitor usage costs, and avoid sending secrets, regulated data, private documents, screenshots, or sensitive image URLs unless you have reviewed and accepted AIsa's privacy, retention, routing, and billing terms.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill clearly requires environment access for `AISA_API_KEY` and network access to send requests to `api.aisa.one`, but it does not declare user-facing permissions or warnings commensurate with those capabilities. This reduces transparency and can cause users or hosting platforms to underestimate that prompts, images, and credentials are being used with an external service.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs users to send chat prompts and vision inputs, including image URLs or base64 image data, to a third-party endpoint without a clear privacy/security warning. In an agent context, users may unknowingly transmit sensitive text, files, screenshots, or internal data to an external provider and potentially onward to downstream model vendors.

Missing User Warnings

Low
Confidence
90% confidence
Finding
The skill tells users to export `AISA_API_KEY` but does not provide any warning about safe credential handling, storage, rotation, or scope. While merely referencing an environment variable is normal, omitting guidance increases the chance of insecure usage such as hardcoding keys, logging them, or reusing overly privileged credentials.

External Transmission

Medium
Category
Data Exfiltration
Content
client = OpenAI(
    api_key=os.environ["AISA_API_KEY"],
    base_url="https://api.aisa.one/v1"
)

response = client.chat.completions.create(
Confidence
88% confidence
Finding
https://api.aisa.one/

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.