ClawHub
Back to skill

Security audit

agency-portal

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent agency portal integration, but it exposes broad tenant, billing, approval, deletion, refund, payment, and admin actions without enough documented safety gates.

Review this skill before installing. Use it only with a trusted MCP server and restrict admin tokens/secrets to operators who are authorized to make billing, tenant, publishing, and account changes. Add or enforce explicit confirmations and audit logging for deletion, refunds, payment confirmation, quota changes, subscription changes, tenant suspension, archiving, and content approval.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger description is extremely broad and covers many sensitive business functions, including account management, billing, approvals, quota changes, refunds, and admin actions, without defining when the skill should or should not activate. In an agent setting, vague activation boundaries can cause the skill to be invoked in the wrong context and lead to unintended handling of sensitive operations or data exposure.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The workflow section enumerates many operations as normal usage patterns but does not specify clear trigger conditions, approval checkpoints, or safety gates for high-risk actions. Because the skill spans login, approvals, renewals, payments, and content actions, lack of trigger specificity increases the chance of overbroad autonomous execution by an agent.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The documentation presents destructive and sensitive operations such as deleting assets, approving/rejecting content, renewing subscriptions, binding/unbinding platform accounts, recording publishes, and other account-affecting actions without user-facing warnings or safety guidance. In a multi-tenant portal/admin context, omission of warnings makes accidental or socially engineered misuse more likely and can result in unauthorized business changes, data loss, or financial impact.

Static analysis

No suspicious patterns detected.