Back to skill

Security audit

AI项目文档生成Skill

Security checks for vulnerabilities and agentic risk

Overview

This is a simple Chinese-language documentation generator with no evidence of hidden data access, persistence, exfiltration, or destructive behavior.

Review the Chinese-language docs before installing, use a pinned or otherwise verified ClawHub installer instead of a mutable `npx` command where possible, and configure platform trigger phrases more specifically if accidental invocation would be disruptive.

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:29
Finding
Unpinned Third-Party Package Execution During Installation## Vulnerability Details **File Location**: `SKILL.md`, line 29 **Vulnerability Type**: Unpinned third-party dependency execution **Risk Level**: Medium **Vulnerable Code Snippet**: ```bash npx clawhub install ai-doc-generator ``` ### Technical Analysis The documented installation command invokes `npx` without specifying a package version or integrity value. Depending on the user's local npm state and registry configuration, `npx` may retrieve and execute the current registry version of `clawhub`. Consequently, the code executed during installation is mutable and lies outside the reviewed project. This creates a supply-chain trust boundary: future package updates, registry-account compromise, package replacement, or a malicious registry configuration could cause a user following the official installation instructions to execute unreviewed code. The audit did not establish that the current `clawhub` package is malicious; the finding concerns the unsafe, unpinned execution mechanism. ### Attack Path 1. An attacker compromises the package publisher, registry entry, distribution infrastructure, or the user's configured npm registry. 2. The attacker publishes or serves a malicious version of the package resolved as `clawhub`. 3. A user follows `SKILL.md` and runs `npx clawhub install ai-doc-generator`. 4. `npx` downloads the mutable package version selected by dependency resolution. 5. Package lifecycle scripts or CLI entry-point code execute with the invoking user's privileges. 6. The malicious code can access resources available to that user and may modify the local environment. ### Impact Assessment Successful exploitation could provide arbitrary code execution under the privileges of the user running the installation command. The potential scope includes readable user files, environment variables, accessible credentials, project files, and network resources available to that account. Further impact would depend on th ...[truncated 192 chars]
Remediation
## Remediation Suggestions 1. Pin the installer to a specifically reviewed version, for example: ```bash npx --yes clawhub@<reviewed-version> install ai-doc-generator ``` 2. Document the official package registry, publisher identity, and source repository so users can verify provenance. 3. Publish and verify package integrity metadata or signatures where the distribution platform supports them. 4. Review the pinned package, including its CLI entry point and npm lifecycle scripts, before recommending execution. 5. Prefer a locally installed, lockfile-controlled dependency or a verified standalone installer over dynamically executing the latest registry release. 6. Advise users not to run the installation command with administrator or root privileges. 7. Add an update process that requires security review before changing the pinned version.
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (8)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The documentation recommends very broad trigger phrases like “生成文档” and “生成日志”, which are common in normal conversation and can cause unintended invocation of the skill. In an agent platform, accidental triggering can lead to unplanned execution paths, confusing behavior, or processing of user content without clear intent.

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
The skill documentation and usage instructions are entirely in Chinese, which implicitly forces a specific language experience without offering users a language or locale choice. Under the policy, language constraints should either be optional or clearly justified as region-specific.

Rp1

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding
The installation command uses `npx clawhub` without pinning an exact version, which can cause users to fetch and execute whatever version is current at install time. This creates a supply-chain risk: a compromised upstream package, typo-squatted replacement, or unexpected breaking update could result in arbitrary code execution on the user's machine.

Natural-Language Policy Violations

Low
Confidence
86% confidence
Finding
该文件从标题到操作说明均仅以中文呈现,未说明是否支持其他语言,也未给出用户可选择语言的机制。按照语言/locale 政策,若技能文档或交互默认强制单一语言而无用户选择,可能构成自然语言策略问题。

Natural-Language Policy Violations

Low
Confidence
94% confidence
Finding
The skill description is presented as a Chinese-only usage and capability statement ('通用型开发者效率工具 - 一键生成专业 README、版本日志') with no indication that users may choose another language or locale. Under the policy rule, natural-language content that imposes a specific language without opt-in can be a locale-policy violation.

Natural-Language Policy Violations

Low
Confidence
91% confidence
Finding
This code file contains natural-language strings that present the skill as a general-purpose developer tool, but the description is written only in Chinese. Because the skill is not documented as region-specific and does not offer user opt-in for language/locale, it may violate language/locale policy expectations.

Natural-Language Policy Violations

Low
Confidence
89% confidence
Finding
The command-line interface text, usage guidance, error messages, and generated section headings are all hard-coded in Chinese. In a general-purpose tool, forcing a single language without opt-in can be a natural-language policy issue.

Natural-Language Policy Violations

Low
Confidence
91% confidence
Finding
This requirements file includes user-facing descriptive text in Chinese, but does not provide an opt-in, alternative language, or justification for a Chinese-only locale. That can violate language/locale policy when users are not explicitly given a choice.

Static analysis

No suspicious patterns detected.