ClawHub
Back to skill

Security audit

create-pr

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed pull-request workflow that can push branches and create PRs, but it keeps those actions tied to the user's PR request and requires approval before pushing.

Install this only if you want your agent to help publish branches and create PRs. Before use, confirm the target repo, branch, provider config, reviewers, auto-complete settings, and work-item links, and do not approve the push unless those details are correct.

SkillSpector

By NVIDIA
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

Medium
Confidence
94% confidence
Finding
The skill declares very broad trigger phrases such as "create a PR", "open a pull request", and especially "ship this branch", and says to use the skill even when the user does not explicitly name it. In an automation context, broad natural-language triggers can cause unintended invocation of a high-impact skill that performs code review, git push, and PR creation actions, increasing the chance of accidental publication workflows being initiated.

Vague Triggers

Low
Confidence
87% confidence
Finding
The usage guidance provides positive invocation examples but insufficiently defines boundaries for when the skill must not activate beyond a brief note about not using it for reviewing an existing PR or local-only git work. Because this skill can push branches and create PRs, unclear trigger boundaries raise the risk of the agent selecting it for loosely related requests, causing unintended side effects in repositories.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.