Back to skill

Security audit

Clausewitz Skill

Security checks for vulnerabilities and agentic risk

Overview

This is a Chinese-language Clausewitz strategy-advice skill with no embedded code or hidden runtime behavior, though its unpinned npx install command carries a supply-chain risk.

Before installing, verify the `dlh365` package source and consider using a pinned, reviewed version or a sandboxed environment. Expect the skill to provide Clausewitz-style strategic advice primarily in Chinese; ask for another language if needed.

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:422
Finding

Unpinned Third-Party Package Execution Through npx

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, line 422
Vulnerability Type: Unpinned third-party dependency execution
Risk Level: Medium

Vulnerable Code:

bash
npx dlh365 add clausewitz-skill

Technical Analysis

The documented installation command invokes dlh365 through npx without specifying an exact package version or integrity constraint. If the package is not already installed locally, npx may retrieve it from the configured package registry and execute its package code, including lifecycle behavior, with the permissions of the invoking user.

Because the dependency is unpinned, the code executed by this command can change after the skill has been reviewed. A compromised maintainer account, malicious future release, registry compromise, or package-name takeover could therefore turn this installation instruction into an arbitrary-code execution path. The document provides no package provenance check, integrity hash, trusted-source restriction, or review step before execution.

Attack Path

  1. An attacker compromises the dlh365 package, its publisher account, or its distribution channel, or otherwise causes a malicious release to be resolved by the package manager.
  2. The attacker publishes a version containing malicious package or lifecycle code.
  3. A user follows the installation command from SKILL.md.
  4. npx resolves and downloads the unpinned package version from the configured registry.
  5. The downloaded package executes with the privileges of the user running the command.
  6. The malicious code can access resources available to that user and may perform additional actions such as modifying files, reading credentials exposed to the process, or downloading further payloads.

Impact Assessment

Successful exploitation could provide arbitrary code execution under the installing user's account. The accessible scope may include user-owned files, project source code, envi ...[truncated 448 chars]

Remediation
View remediation

Remediation Suggestions

  • Pin dlh365 to an exact version that has been reviewed, rather than allowing npx to resolve a mutable release.
  • Verify the package publisher, repository, release provenance, signatures, and registry metadata before recommending execution.
  • Use a lockfile or another integrity-verification mechanism that records and enforces expected package hashes.
  • Prefer downloading the package artifact for inspection before executing it, especially when installation instructions are intended for broad distribution.
  • Run the installation in a sandbox, container, or minimally privileged account without access to sensitive credentials.
  • Configure the package manager to use an approved registry and disable or restrict lifecycle scripts where operationally possible.
  • Document the exact audited package version and update it only after reviewing subsequent releases.
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The entire skill description and examples are written in Chinese, with no indication that users may choose another language or that the Chinese-only presentation is a justified regional constraint. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

The skill instructs users to run npx dlh365 add clausewitz-skill without pinning an exact package version. This can cause execution of whatever package version is current at install time, creating a supply-chain risk if the package is updated maliciously, compromised, or changed incompatibly. Because the command is presented as the installation method, users may execute remote code from npm with little scrutiny.

Content

No source excerpt is available for this finding.