ClawHub
Back to skill

Security audit

api-connect-hub

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent API integration helper with no hidden executable files, but it requests broad command/write authority and long-lived credential handling without clear enough consent boundaries.

Install only if you intend to let the agent help manage real API integrations. Use least-privilege credentials, avoid production tokens until tested, require explicit approval before shell commands, connector installs/publishes, OAuth refresh setup, webhook registration, or data sync runs, and review generated configs before use.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The skill’s invocation guidance is vague and does not clearly define when it should be used, what inputs are in scope, or what actions require explicit user confirmation. In a skill that advertises read/exec/write capabilities and API orchestration, ambiguous triggering can cause the agent to over-apply the skill, execute commands, or handle credentials in contexts the user did not intend.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill describes token management, credential storage locations, Vault usage, and command execution, but does not provide a prominent warning that these actions may access secrets, modify local state, or invoke external systems. Because the skill includes exec capability and is designed for API integration, users may unknowingly expose sensitive data or trigger side effects without informed consent.

Static analysis

No suspicious patterns detected.