Back to skill

Security audit

data-export-check

Security checks for vulnerabilities and agentic risk

Overview

The skill itself is a simple local compliance checklist, but its install instructions use mutable remote sources that users should review before running.

Review or pin the exact repository commit/package version before installing. The local script appears limited to checklist evaluation, but avoid the documented one-command global install unless you trust the current upstream package and repository state.

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:81
Finding
Unpinned Third-Party Package and Repository Installation## Vulnerability Details **File Location**: `SKILL.md`, lines 81-86 **Vulnerability Type**: Unpinned and mutable third-party installation sources **Risk Level**: Medium ```bash # One-command installation (skills CLI) npx skills add zhaoxinghua09-cell/agent-skills -g # Or manually: clone and copy this skill into the Agent skill directory git clone https://github.com/zhaoxinghua09-cell/agent-skills.git ``` ### Technical Analysis The documented one-command installation invokes `npx skills` without specifying an exact package version or integrity digest. Depending on the local npm configuration and cache state, `npx` can download and execute the currently resolved version of the package. The executed dependency is not included in this audited artifact, so its behavior cannot be verified by this review. The alternative installation procedure clones a mutable repository reference without specifying a reviewed commit hash or signed release. Consequently, the repository content installed by a user can differ from the content that was audited. This creates a supply-chain trust gap: compromise of the package registry account, npm package, source repository, maintainer account, or upstream release process could cause users to retrieve altered code or Skill instructions. ### Attack Path 1. An attacker compromises or gains publishing control over the package resolved by `npx skills`, or modifies the referenced source repository. 2. The attacker publishes a malicious package version or commits altered scripts and Skill instructions to the repository's default branch. 3. A user follows the installation commands in `SKILL.md`. 4. In the `npx` path, the mutable package is retrieved and its code can execute with the invoking user's privileges. 5. In the Git path, mutable repository content is copied into the user's Agent skill directory and may subsequently be loaded or executed as a trusted Skill. 6. The malicious dependency or Skill can access resources available to th ...[truncated 796 chars]
Remediation
## Remediation Suggestions 1. Pin the `npx` package to an exact reviewed version rather than resolving the latest available release: ```bash npx skills@<reviewed-version> add zhaoxinghua09-cell/agent-skills@<reviewed-version> -g ``` 2. Document the expected npm package integrity digest and provide a verification procedure before execution. 3. Prefer a locally installed, lockfile-controlled CLI over dynamically downloading and executing a package through `npx`. 4. Pin the Git installation to a specific reviewed commit: ```bash git clone https://github.com/zhaoxinghua09-cell/agent-skills.git cd agent-skills git checkout --detach <reviewed-commit-sha> ``` 5. Publish signed tags or release artifacts and require users to verify signatures or checksums before installation. 6. Avoid global installation by default. Install into a dedicated, least-privileged Agent directory or isolated environment. 7. Ensure updates undergo the same security review and require explicit user approval before changing the pinned version or commit.
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (6)

Rp1

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding
The skill instructs users to run `npx skills add ...` without pinning an exact package version, which can cause execution of whatever package version is current at install time. If the upstream package is compromised, typosquatted, or updated with malicious behavior, users may execute unreviewed code during installation or skill acquisition.

Natural-Language Policy Violations

Medium
Confidence
89% confidence
Finding
The manifest sets the category value to Chinese text ("AI 治理"), which signals a fixed language choice in user-facing metadata. There is no indication elsewhere in the file that the skill offers language selection or that this locale restriction is intentionally documented as region-specific.

Natural-Language Policy Violations

Medium
Confidence
96% confidence
Finding
The user-facing strings, checklist items, and output messages are entirely in Chinese, and the file provides no option to select another language or any note that the skill is intentionally limited to a Chinese-language/regulatory context. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy violation.

Natural-Language Policy Violations

Low
Confidence
81% confidence
Finding
The heading begins in Chinese and presents the document bilingually, but the substantive attestation content on subsequent lines is primarily in Chinese. This can constitute a language/locale policy issue because the file does not indicate user choice or opt-in for the non-English primary content.

Natural-Language Policy Violations

Low
Confidence
79% confidence
Finding
The file prominently mixes Chinese-first labeling and domain description without stating that users may choose their preferred language or locale. Under the policy, language constraints should either be optional or clearly justified; this README does not explicitly provide that opt-in choice.

Natural-Language Policy Violations

Low
Confidence
89% confidence
Finding
The natural-language instructions, usage guidance, warnings, and disclaimers are entirely in Chinese, while only the display name includes a brief English label. This can amount to a language-policy issue if users are not given an explicit language opt-in or an explanation that the skill is intended only for Chinese-speaking or China-specific compliance contexts.

Static analysis

No suspicious patterns detected.