Back to skill

Security audit

Find Docs

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent documentation lookup helper, but it asks agents to install or run a mutable third-party CLI broadly for many technical tasks.

Review before installing. Prefer running a pinned ctx7 version in an isolated environment, avoid global installation, and make sure documentation queries do not include credentials, personal data, proprietary code, or sensitive internal details.

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 CLI Installation and Execution

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 29–39
Vulnerability Type: Supply-chain risk caused by installing and executing an unpinned dependency
Risk Level: Medium

Vulnerable code:

bash
Make sure the CLI is up to date before running commands:

```bash
npm install -g ctx7@latest

Or run directly without installing:

bash
npx ctx7@latest <command>
text

### Technical Analysis

The skill explicitly instructs the agent to download and execute the mutable `latest` release of the `ctx7` npm package. Neither command identifies a reviewed, immutable package version or verifies the downloaded artifact's integrity.

`npx ctx7@latest` may download package content and execute its binaries immediately. Similarly, `npm install -g ctx7@latest` installs the current registry-selected release globally and may run npm lifecycle scripts during installation. Consequently, the code actually executed can change after the skill has been audited without any modification to `SKILL.md`.

This creates a supply-chain trust boundary around the npm registry, the package publisher account, package maintainers, and the local npm registry configuration. A compromised publisher account, malicious future release, registry redirection, or compromised configured registry could cause attacker-controlled code to run when the skill workflow is followed.

No evidence in the audited file establishes that the current `ctx7` package is malicious. The confirmed issue is the unsafe use of an unpinned, automatically changing dependency execution mechanism.

### Attack Path

1. An attacker compromises the `ctx7` publishing process or publisher account, publishes a malicious release selected by the `latest` tag, or causes the environment to resolve the package through a malicious npm registry.
2. The agent follows the skill instruction and runs `npm install -g ctx7@latest` or `npx ctx7@latest`.
3. npm retrieves the
...[truncated 1245 chars]
Remediation
View remediation

Remediation Suggestions

  1. Replace ctx7@latest with an exact, reviewed version, such as ctx7@X.Y.Z. Do not use mutable tags or semver ranges in executable instructions.
  2. Record and verify the expected package integrity hash before execution. Where feasible, manage the CLI through a project lockfile and use npm ci so dependency resolution is reproducible.
  3. Avoid global installation. Install the pinned package in a dedicated project or isolated tool directory to limit modifications to the user's global environment.
  4. Disable lifecycle scripts during installation when they are not required, for example with npm install --ignore-scripts, and separately review any scripts that must run.
  5. Require the official npm registry explicitly and validate that local or environment-level npm configuration has not redirected package resolution to an untrusted registry.
  6. Execute the documentation CLI in a sandbox or restricted container with minimal filesystem access, no unnecessary credentials, and constrained outbound network access.
  7. Establish a controlled update process: review new releases, verify publisher and provenance information, test the package in isolation, and update the pinned version only after approval.
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
94% confidence
Finding

The activation guidance is extremely broad, effectively telling the agent to invoke this skill for most technical questions and code-generation tasks. Overly broad routing increases unnecessary exposure to external tooling, raises the chance that sensitive user prompts are sent as documentation queries, and can create avoidable supply-chain or data-handling risk even though the skill does warn against including secrets.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.