ClawHub
Back to skill

Security audit

vibo-mcp

Security checks for vulnerabilities and agentic risk

Overview

The skill appears purpose-built for Vibo event music management, but it asks users to put account credentials or captured session tokens into local MCP configuration/session files without enough security warning.

Review before installing. This skill may be useful if you want an agent to manage Vibo event music, but treat the Vibo password, access token, refresh token, and ~/.vibo-mcp/session.json as secrets. Avoid committing MCP config files, restrict file permissions, and revoke or rotate credentials if they are exposed. Keep confirmation mode enabled for write actions.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

MCP Config Access

High
Category
Agent Snooping
Confidence
95% confidence
Finding

The setup directs users to place plaintext credentials into project-level or user-level MCP config files, including under ~/.claude/mcp.json, which can expose secrets to local compromise, backups, source control mistakes, or other tools that read configuration. Because the skill specifically names credential-bearing env vars and a common agent config path, it materially increases the chance of unsafe secret handling rather than merely referencing configuration in the abstract.

Content

Scanner excerpt · SKILL.md (reported line 17)May include surrounding context.

Option A — npx (recommended)

Add to .mcp.json in your project or ~/.claude/mcp.json:

json
{

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill explicitly instructs users to use captured access and refresh tokens from an active web session and to persist browser-captured session data locally, but it does not include a clear warning that these credentials are highly sensitive and equivalent to account access. In a skill context, this increases the risk of credential theft, accidental disclosure in config files, logs, screenshots, or prompt history, especially because the skill is meant to be copied into local MCP configuration and used by non-security-focused users.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 107)May include surrounding context.

md
## Response shape (`view`)

**Eight of this server's 42 tools take `view: "compact" | "full"`**, and on every
one of them **`compact` is the DEFAULT**. You get the slim rung without asking.

They are exactly the eight reads whose GraphQL document asks Vibo for media:

Static analysis

No suspicious patterns detected.