Back to skill

Security audit

Console Patrol

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed web console scanner, but users should limit route scanning and install its dependencies in a sandboxed environment.

Install and run this only in a least-privileged, isolated environment. Prefer approved test routes, avoid sensitive production data, use screenshots only when needed, and pin or vet dependencies before relying on it in CI or shared agent environments.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:20
Finding

Unpinned Third-Party Package Installation

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 20–25 and 149–156
Vulnerability Type: Unpinned and externally retrieved dependencies
Risk Level: Medium

Vulnerable Code

bash
pip install console-patrol
playwright install chromium

The same installation behavior is presented programmatically later in the file:

python
import subprocess
subprocess.run(["pip", "install", "console-patrol"], check=True)
subprocess.run(["playwright", "install", "chromium"], check=True)

Technical Analysis

The Skill instructs an agent to retrieve and install console-patrol from the package index without specifying an audited version, validating cryptographic hashes, using a locked transitive dependency set, or restricting resolution to a documented trusted source. It also invokes Playwright to download a Chromium binary without pinning or integrity controls in the Skill.

Because neither the third-party package implementation nor the downloaded browser artifact is included in the audited project, their effective behavior cannot be verified from this repository. A future compromised package release, malicious transitive dependency, dependency-resolution attack, or compromised distribution source could introduce attacker-controlled code after the Skill itself has been reviewed.

Attack Path

  1. An attacker compromises a future console-patrol release, one of its transitive dependencies, or a package distribution source used by the environment.
  2. An agent follows the instructions in SKILL.md and runs the unpinned installation command.
  3. Package installation or build hooks execute attacker-controlled code, or malicious package code is installed for execution during a subsequent scan.
  4. The code runs with the permissions and environmental access of the agent process.
  5. The payload may read accessible files or credentials, alter the environment, monitor browser activity, or communicate dat ...[truncated 738 chars]
Remediation
View remediation

Remediation Suggestions

  1. Pin console-patrol, Playwright, and all transitive dependencies to versions that have been reviewed and tested.
  2. Maintain a lock file or fully pinned requirements file with cryptographic hashes, and install with pip's --require-hashes option.
  3. Use a controlled package index or internal artifact repository containing approved artifacts rather than resolving mutable packages directly from public sources.
  4. Pin and validate the Playwright browser artifact through a controlled build process.
  5. Generate and review a software bill of materials, and continuously scan dependencies for known vulnerabilities and unexpected ownership or release changes.
  6. Perform installation and page scans in an ephemeral, sandboxed, least-privileged environment with minimal filesystem, credential, and network access.
  7. Avoid running package installation as an administrator or root user.
  8. Prefer including or vendoring the reviewed implementation where licensing and maintenance requirements permit, so the executed code corresponds to the audited artifact.
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
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger phrases are broad enough to match generic requests like 'audit a web app' or 'find issues in the application,' which can cause this skill to activate when the user did not specifically intend a console scan. Unintended invocation matters because the skill can drive a browser against application pages and potentially perform extra scanning actions beyond the user's expected scope.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill notes features like screenshots and route scanning but does not clearly warn that execution may open application pages, traverse routes, and capture potentially sensitive page contents. In an agent setting, that omission can lead to user surprise, privacy issues, or scanning of internal/admin pages without sufficiently explicit consent.

Content

No source excerpt is available for this finding.