ClawHub
Back to skill

Security audit

Api Connect Hub Free

Security checks across malware telemetry and agentic risk

Overview

The skill is not overtly malicious, but it needs review because it handles credentials and API actions while using broad triggers and inconsistent free-edition guidance.

Review this before installing if you expect webhook management or OAuth2 refresh support: the free-edition text conflicts on both. Use it only for explicit, user-directed API connector and credential-template tasks, and avoid letting it automatically edit scripts or handle real tokens without checking the generated changes.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
88% confidence
Finding
The skill advertises Webhook configuration/use cases in the manifest and description, but later states Webhook management is unavailable in the free edition. This inconsistency can cause unintended invocation for webhook-related tasks and may lead users or agents to generate unsafe or nonfunctional webhook handling guidance under false assumptions.

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The document says the free edition does not support automatic OAuth2 refresh, yet the retry rules instruct agents to refresh tokens and retry on 401. This contradiction can cause an agent to invent or attempt token refresh flows that are unsupported, risking broken authentication logic, accidental mishandling of refresh tokens, or insecure custom implementations.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The activation description is broad enough to match many generic API integration, webhook, and system connection requests. Over-broad triggering increases the chance the skill is invoked in contexts it was not designed for, which is risky here because it has exec capability and can influence credential handling and outbound API usage.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The capability coverage keywords include very generic terms like 'when', 'Webhook', '系统连接', and other broad fragments that could match routine user requests too loosely. This expands invocation scope unnecessarily and may cause the skill to handle sensitive integration tasks without sufficient user specificity or safety checks.

VirusTotal

56/56 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.