ClawHub
Back to skill

Security audit

ReviewEvo

Security checks across malware telemetry and agentic risk

Overview

This skill is a local code-review helper whose repository reads and local learning file are disclosed and aligned with its purpose, but users should manage the saved review notes carefully.

Install this only for repositories where you are comfortable with the agent reading code and git history. Before committing, inspect .review-evo/learnings.md and add .review-evo/ to .gitignore if the notes should remain local. For branch reviews, provide normal branch names and avoid shell-like input.

SkillSpector

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

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill explicitly instructs the agent to create and append to `.review-evo/learnings.md`, which modifies the user's repository as part of normal execution. Because this write happens after review flow and is presented as a default behavior rather than an opt-in step with clear warning, it can cause unexpected workspace changes, pollute repos, and potentially commit local analysis artifacts or sensitive notes if the user is not aware.

Ssd 3

Medium
Confidence
90% confidence
Finding
The skill advertises a persistent knowledge base that 'learns your codebase over time,' which implies retention of repository-derived findings across runs. Even without exfiltration, persisting review conclusions, code patterns, and user context locally can accumulate sensitive operational details and create unintended long-term storage of proprietary or security-relevant information.

Ssd 3

Medium
Confidence
97% confidence
Finding
This section directs the agent to append cumulative project profile, hotspots, recurring patterns, resolved issues, and open risks into a local file for future reuse. That creates durable storage of potentially sensitive repository structure, weaknesses, and developer activity patterns, increasing exposure if the repo is shared, committed, or accessed by other tools or users.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal