ClawHub
Back to skill

Security audit

jules

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Jules CLI reference whose repository-changing commands are disclosed and aligned with its purpose, but users should review remote patches before applying them.

Before using `--apply` or `teleport`, inspect the fetched patch, confirm your git status and target branch, and prefer a clean branch or disposable checkout for remote agent-generated changes.

SkillSpector

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 (2)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill explicitly recommends commands that can automatically modify the current repository (`jules remote pull --session <id> --apply`) or clone and apply changes (`jules teleport <id>`) without an adjacent warning about workspace mutation, file creation, or the trust boundary of remotely generated patches. Because Jules is an autonomous coding agent producing code asynchronously, applying its output directly increases the risk of unintended, unsafe, or malicious code being introduced into the user's checkout.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The documentation instructs users to fetch and directly apply remotely generated patches to the local working tree, including a one-shot teleport flow that may clone and apply changes automatically, but it does not warn about local modification, patch trust, branch safety, or review requirements. In a coding-agent CLI context, this is dangerous because users may treat cloud-generated changes as trusted and unintentionally overwrite local work or introduce malicious or unsafe code into a repository.

Static analysis

No suspicious patterns detected.