ClawHub
Back to skill

Security audit

Revibe Codes

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent code-analysis skill, but it can send and store repository code with Revibe from broad or silent triggers without strong per-run consent.

Install only if you are comfortable letting Revibe analyze and retain the target repository code. Be especially careful with private, proprietary, regulated, or client code; prefer explicit repository URLs, avoid broad allowed-tools grants unless you trust the service, and review Revibe's privacy and retention terms before use.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger conditions are broad enough to activate on common requests like 'analyze' or 'understand this project,' which can cause the skill to run and transmit repository information to a third-party service without sufficiently explicit user intent. In a security-sensitive environment, broad activation increases the chance of accidental data exposure, especially when the current working directory is auto-detected as the target repository.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
97% confidence
Finding

The skill explicitly sends repository identifiers to an external service and states that source code is stored in cloud storage for analysis and re-analysis. Even though the privacy note discloses this behavior, it still creates a real data exfiltration risk for private, proprietary, or sensitive repositories if users invoke the skill without fully appreciating the transmission and retention implications.

Content

Scanner excerpt · SKILL.md (reported line 32)May include surrounding context.

md
This skill sends your repository's GitHub URL to revibe.codes for analysis. Source code is stored securely in Google Cloud Storage to enable features like code exploration and re-analysis. Only you can access your uploaded projects. If you're working with private or sensitive repositories, review revibe.codes/privacy before proceeding.

**Tip:** To reduce permission prompts, you can optionally add `Bash(curl *revibe.codes*)` to your allowed tools via `/allowed-tools`.

## Behavior

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The drill-down example says the skill will 'render Mermaid as ASCII', but the Diagram Rendering section specifies writing Mermaid to an HTML file and opening it in a browser. This is an active contradiction between documentation sections about what the skill does when showing diagrams.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.