ClawHub
Back to skill

Security audit

Parallel Code Review

Security checks for vulnerabilities and agentic risk

Overview

This is a code-review helper that can send selected repository diffs to external review CLIs, but its behavior is disclosed, scoped to review work, and includes meaningful safeguards.

Install this if you want an automated code-review closeout helper and are comfortable with changed repository content being sent to the selected reviewer provider. Before running it on sensitive work, choose an explicit mode/base or commit, use --dry-run to confirm scope, disable web search if not needed, and avoid including secrets in diffs or prompt files.

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 (2)

Vague Triggers

High
Confidence
94% confidence
Finding
The trigger phrases are broad enough to activate on common user requests like "review" or "code review," which can cause the skill to run unintentionally in contexts where the user did not mean to invoke this specific behavior. Because the skill defaults to reviewing all unmerged commits on the current branch, accidental activation can expose more repository content and consume more agent effort than the user expected.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README says the skill defaults to reviewing all commits on the current branch not yet on upstream, but the description and trigger behavior do not prominently warn users about that scope at invocation time. This can lead to unintended analysis of a large or sensitive set of changes when a user issues a casual review request, creating privacy, cost, and operational-risk concerns.

Static analysis

No suspicious patterns detected.