ClawHub
Back to skill

Security audit

model-intel-check(zjh)

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches its model-benchmarking purpose, but it needs review because it uses local API credentials and includes WAF-bypass User-Agent guidance.

Install only if you are comfortable with the agent reading or using model-provider API credentials for the endpoint you choose. Prefer manually supplying a temporary API key via environment variables, review provider terms before running high-concurrency benchmarks, and avoid relying on the WAF-bypass guidance unless you have authorization from the provider or intermediary.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill instructs reading provider credentials from `~/.kimi-code/config.toml` and handling `api_key` values without an explicit consent gate or warning, which risks unauthorized access to secrets beyond what the user intended to share. Because the skill also uses those credentials for outbound network requests, accidental activation or over-broad execution could expose or misuse sensitive provider access.

Ssd 2

Medium
Confidence
98% confidence
Finding
The skill explicitly recommends using a spoofed `User-Agent` to bypass a WAF-style block, which is an evasion technique intended to circumvent an intermediary's access controls. Even if framed as reliability troubleshooting, this normalizes bypassing defensive mechanisms and could violate provider policies or be repurposed to evade monitoring and restrictions.

Static analysis

No suspicious patterns detected.