ClawHub
Back to skill

Security audit

jules

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only skill for using the Jules CLI, and its repository-changing commands are disclosed and aligned with that purpose.

Install only if you intend to use Jules for cloud-based coding work. Before using `--apply` or `teleport`, review the patch, confirm the session and repository, and prefer applying in a clean branch or disposable checkout when possible.

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
  • 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
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill explicitly recommends commands that can modify the current checkout (jules remote pull --session <id> --apply) or clone and apply changes automatically (jules teleport <id>) without any adjacent warning to review diffs, verify target repos, or protect the local workspace first. Because this skill is specifically about delegating code changes to a remote autonomous agent, encouraging direct application of returned patches increases the chance of unreviewed, workspace-modifying, or supply-chain-impacting changes being introduced.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The documentation advertises commands that can directly modify the local working tree by applying remote-generated patches or cloning and applying changes, but it does not clearly warn users about the trust boundary or the risk of unintended code modification. In this skill context, the tool is specifically designed to fetch and apply AI-generated changes from remote sessions, which makes silent or lightly documented workspace modification more dangerous because users may treat the output as routine automation and apply unreviewed changes.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.