Back to skill

Security audit

课时核销核对(免费版)

Security checks for vulnerabilities and agentic risk

Overview

The free checker itself is local and purpose-aligned, but its documentation includes an optional paid-upgrade flow that asks an agent to run unpinned external installer code and install another skill.

The bundled free checker can be used as a local table-checking script. Do not let an agent run the paid-upgrade installer or purchase/install the full version unless you manually verify the package source, pin or inspect the exact version, understand the payment flow, and are comfortable adding new skill instructions to the agent environment.

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:66
Finding
Execution of an Unpinned Third-Party Package Through npx## Vulnerability Details **File Location**: `SKILL.md`, line 66 **Vulnerability Type**: Unpinned third-party dependency execution **Risk Level**: Medium **Vulnerable Code**: ```bash npx -y @alipay/agent-payment@latest install ``` ### Technical Analysis The optional installation instructions direct the agent to download and execute the `latest` release of a third-party npm package. The `latest` tag is mutable, so the code executed in the future may differ from the version available when this Skill was audited. The `-y` option also suppresses the normal package-installation confirmation. The surrounding instructions subsequently direct the agent to read and follow instructions from the newly installed `skillpay-onboarding` Skill. Those external instructions and their effective code are not included in this project and therefore are outside the reviewed trust boundary. This is a supply-chain risk rather than evidence that the package is currently malicious. The vulnerable path is optional and is not invoked by the bundled lesson-hour checking runtime. Nevertheless, compromise of the package, publisher account, npm distribution path, or mutable release tag could turn the documented installation procedure into arbitrary local code execution. ### Attack Path 1. An attacker compromises the npm package, its publisher account, or another part of its distribution chain. 2. The attacker publishes a malicious release and assigns it to the mutable `latest` tag. 3. A user or agent follows the installation instructions in `SKILL.md`. 4. `npx -y` retrieves the current malicious release without an interactive installation prompt. 5. npm package lifecycle or executable code runs with the permissions of the invoking user. 6. The agent may then load and follow additional attacker-controlled instructions from the installed Skill. ### Impact Assessment Successful exploitation could provide arbitrary code execution with the privileges ...[truncated 529 chars]
Remediation
## Remediation Suggestions 1. Replace `@latest` with an explicitly reviewed and immutable package version. 2. Verify the downloaded package against a documented integrity digest or lockfile before execution. 3. Remove `-y` so users receive a confirmation opportunity before package execution. 4. Require explicit user approval immediately before running the installation command. 5. Document the expected npm publisher, package version, integrity value, and official registry source. 6. Inspect the installed Skill and its scripts before loading or following its instructions. 7. Prefer distributing reviewed installation logic within the audited package instead of dynamically executing mutable external code. 8. Run any unavoidable installer in a sandbox with minimal filesystem, credential, network, and process privileges.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (7)

Ae1

High
Category
analysis-evasion
Content
node scripts/run.mjs --sample # 先看样例
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

Ae1

High
Category
analysis-evasion
Content
node scripts/run.mjs --sample # 先看样例
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The English description claims 'No payment, no registration, no API key, no network,' yet the body includes explicit purchase and installation instructions for a paid product. This inconsistency can mislead users and downstream agents about the skill's behavior and safety properties, increasing the chance they follow risky payment or install prompts under false assumptions.

Description-Behavior Mismatch

Medium
Confidence
89% confidence
Finding
The skill presents itself as a free, local-only checker but also embeds a workflow to purchase and install a separate paid skill. That cross-sell path changes the trust boundary by steering users from a supposedly self-contained local tool into executing external installation and payment steps, which can be abused for social engineering or unsafe tool expansion.

Natural-Language Policy Violations

Medium
Confidence
92% confidence
Finding
The embedded instructions explicitly tell the user to send a Chinese prompt verbatim to their AI assistant, and the skill otherwise presents its operational guidance primarily in Chinese. There is no explicit user choice of language or justification that the workflow must be Chinese-only, which creates a natural-language locale policy concern.

Rp1

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding
The skill instructs the user to run `npx -y @alipay/agent-payment@latest install`, which fetches and executes remote code at install time without a pinned version. This creates a supply-chain execution risk: if the package is compromised or a breaking update is published, users may run attacker-controlled code simply by following the skill's instructions.

Natural-Language Policy Violations

Low
Confidence
89% confidence
Finding
The JSON contains only Chinese-language tabular content, which implies a fixed language/locale with no opt-in or alternative. Under the policy, locale-specific language is a finding when the skill does not offer user choice or justify the constraint.

Static analysis

No suspicious patterns detected.